ENHANCED PARKWAY STUDY: PHASE 2 CONTINUOUS FLOW INTERSECTIONS. Final Report

Transcription

1 Preparedby:

2

3 ENHANCED PARKWAY STUDY: PHASE 2 CONTINUOUS FLOW INTERSECTIONS Final Report Prepared for Maricopa County Department of Transportation Prepared by

4

5 TABLE OF CONTENTS Page EXECUTIVE SUMMARY ES-1 STUDY OVERVIEW ES-1 STUDY METHODOLOGY ES-1 SUMMARY OF RESULTS AND CONCLUSIONS ES-4 1. INTRODUCTION AND STUDY OVERVIEW 1 2. GENERAL ANALYSIS APPROACH AND METHODOLOGY 4 BASE NETWORK CONFIGURATION 4 MLT NETWORK CONFIGURATION 4 SINGLE POINT URBAN INTERCHANGE (SPUI) CONFIGURATION 9 CONTINUOUS FLOW INTERSECTION CONFIGURATION 10 TRAFFIC ANALYSIS METHODOLOGY 12 TRAFFIC VOLUMES 12 SIMULATION OF MLT AND CONTINUOUS FLOW INTERSECTIONS 13 ALTERNATIVE ANALYSIS SCENARIOS 17 TRAFFIC SIGNAL TIMING FOR THE ANALYSIS AND COMPARISON OF ALTERNATIVES 18 PERFORMANCE ANALYSIS METRICS SUMMARY OF TRAFFIC VOLUME, SIGNAL TIMING, AND TRAFFIC SPEED INPUT VALUES 25 TRAFFIC VOLUME INPUT DATA 25 BASE CAPACITY TRAFFIC VOLUMES 25 MLT CAPACITY VOLUMES 25 CFI CAPACITY VOLUMES 25 TRAFFIC SIGNAL TIMING INPUT VALUES 25 TRAFFIC SPEEDS ANALYSIS RESULTS 33 BASE CAPACITY TRAFFIC VOLUME SCENARIO 33 Aggregate Results for Three Study Intersections with Base Capacity Volumes 34 Parkway-Parkway Intersection (Mile 6) with Base Capacity Volumes 34 Parkway-Major Arterial Intersection (Mile 10) with Base Capacity Volumes 37 Parkway-Minor Arterial Intersection (Mile 7) with Base Capacity Volumes 40 MLT CAPACITY TRAFFIC VOLUME SCENARIO 43 Aggregate Results for Three Study Intersections with MLT Capacity Volumes 43 Parkway-Parkway Intersection (Mile 6) with MLT Capacity Volumes 44 Parkway-Major Arterial Intersection (Mile 10) with MLT Capacity Volumes 47 Parkway-Minor Arterial Intersection (Mile 7) with MLT Capacity Volumes 49 PARKWAY PARKWAY INTERSECTION (Mile 6) ANALYSIS ASSUMING UNIFORM VOLUMES 51 CFI CAPACITY TRAFFIC VOLUME SCENARIO 54 Phase 2 Continuous Flow Intersections

6 TABLE OF CONTENTS Page 4. ANALYSIS RESULTS (Continued) 33 CFI UNIFORM CAPACITY TRAFFIC VOLUME SCENARIO 57 Comparison of the CFI and SPUI Operations with CFI Uniform Capacity Volumes 60 RIGHT-OF-WAY CONSIDERATIONS 62 CFI ACCESS CONSIDERATIONS SUMMARY OF RESULTS AND CONCLUSIONS 65 Phase 2 Continuous Flow Intersections

7 LIST OF EXHIBITS Page EXHIBIT ES-1 ILLUSTRATION OF A CONTINUOUS FLOW INTERSECTION EXHIBIT ES-2 EXAMPLE OF A TYPICAL MLT INTERSECTION EXHIBIT ES-3 COMPARISON OF CFI TO MLT DESIGN BY ANALYSIS SCENARIO EXHIBIT ES-4 COMPARISON OF SPUI TO CFI DESIGN BY ANALYSIS SCENARIOS ES-2 ES-4 ES-6 ES-7 EXHIBIT 1-1 EXAMPLE OF A CONTINUOUS FLOW INTERSECTION 2 EXHIBIT 1-2 EXAMPLE OF A TYPICAL MLT INTERSECTION 3 EXHIBIT 2-1 GENERAL ROADWAY SCHEME FOR THE ANALYSIS 6 EXHIBIT 2-2 BASE NETWORK INTERSECTION LANE CONFIGURATION, LANE USE, AND INTERSECTION CONTROL ASSUMPTIONS 7 EXHIBIT 2-3 MLT NETWORK LANE CONFIGURATION, LANE USE, AND INTERSECTION CONTROL ASSUMPTIONS 8 EXHIBIT 2-4 ASSUMED SPUI CONFIGURATION FOR THE PARKWAY-PARKWAY INTERSECTION 9 EXHIBIT 2-5 CFI INTERSECTION SCHEMATIC CONFIGURATIONS AND TRAFFIC CONTROL 11 EXHIBIT 2-6 SIMULATION IMAGE OF AN MLT INTERSECTION 14 EXHIBIT 2-7 SIMULATION IMAGE OF A CFI ON THE BASE NETWORK 15 EXHIBIT 2-8 SIMULATION IMAGE OF A CFI ON THE MLT NETWORK 16 EXHIBIT 2-9 COMPARISON OF TRAFFIC MOVEMENTS THROUGH A STANDARD INTERSECTION AND AN MLT INTERSECTION 21 EXHIBIT 2-10 STUDY METHOD OF DELAY COMPARISON FOR AN MLT INTERSECTION 22 EXHIBIT 2-11 TRAFFIC MOVEMENTS THROUGH A CONTINUOUS FLOW INTERSECTION 23 EXHIBIT 2-12 STUDY METHOD OF DELAY COMPARISON FOR A CONTINUOUS FLOW INTERSECTION 24 EXHIBIT 3-1 BASE CAPACITY INPUT TRAFFIC VOLUMES 27 EXHIBIT 3-2 MLT CAPACITY INPUT TRAFFIC VOLUMES 28 EXHIBIT 3-3 PARKWAY-PARKWAY INTERSECTION ANALYSIS INPUT MLT UNIFORM CAPACITY VOLUMES 28 EXHIBIT 3-4 CFI CAPACITY INPUT TRAFFIC VOLUMES 29 EXHIBIT 3-5 PARKWAY-PARKWAY INTERSECTION ANALYSIS INPUT CFI UNIFORM CAPACITY VOLUMES 29 EXHIBIT 3-6 TYPICAL TRAFFIC SIGNAL TIMING BY ROADWAY TYPE 30 EXHIBIT 3-7 TRAFFIC SIGNAL TIMING FOR THE PARKWAY-PARKWAY INTERSECTION ANALYSES 31 EXHIBIT 3-8 INPUT ROADWAY TRAVEL SPEED BY ROADWAY TYPE 32 EXHIBIT 4-1 EXHIBIT 4-2 AGGREGATE INTERSECTION PERFORMANCE MEASURES WITH BASE CAPACITY VOLUMES 34 INTERSECTION PERFORMANCE MEASURES FOR THE PARKWAY- PARKWAY (MILE 6) INTERSECTION WITH BASE CAPACITY VOLUMES 35 Phase 2 Continuous Flow Intersections

8 LIST OF EXHIBITS Page EXHIBIT 4-3 PARKWAY-PARKWAY INTERSECTION (MILE 6) DELAY PER MOVEMENT BY INTERSECTION DESIGN ALTERNATIVE (Base Capacity Volumes) 36 EXHIBIT 4-4 INTERSECTION PERFORMANCE MEASURES FOR THE PARKWAY- MAJOR ARTERIAL (MILE 10) INTERSECTION WITH BASE CAPACITY VOLUMES 38 EXHIBIT 4-5 PARKWAY-MAJOR ARTERIAL INTERSECTION (MILE 10) DELAY PER MOVEMENT BY INTERSECTION DESIGN ALTERNATIVE (Base Capacity Volumes) 39 EXHIBIT 4-6 INTERSECTION PERFORMANCE MEASURES FOR THE PARKWAY- MINOR ARTERIAL (MILE 7) INTERSECTION WITH BASE CAPACITY VOLUMES 40 EXHIBIT 4-7 PARKWAY-MINOR ARTERIAL INTERSECTION (MILE 7) DELAY PER MOVEMENT BY INTERSECTION DESIGN ALTERNATIVE (Base Capacity Volumes) 42 EXHIBIT 4-8 AGGREGATE INTERSECTION PERFORMANCE MEASURE WITH MLT CAPACITY VOLUMES 44 EXHIBIT 4-9 INTERSECTION PERFORMANCE MEASRUES FOR THE PARKWAY- PARKWAY (MILE 6) INTERSECTION WITH MLT CAPACITY VOLUMES 45 EXHIBIT 4-10 PARKWAY-PARKWAY INTERSECTION (MILE 6) DELAY PER MOVEMENT BY INTERSECTION DESIGN ALTERNATIVE (MLT Capacity Volumes) 46 EXHIBIT 4-11 INTERSECTION PERFORMANCE MEASURES FOR THE PARKWAY- MAJOR ARTERIAL (MILE 10) INTERSECTION WITH MLT CAPACITY VOLUMES 47 EXHIBIT 4-12 PARKWAY-MAJOR ARTERIAL INTERSECTION (MILE 10) DELAY PER MOVEMENT BY INTERSECTION DESIGN ALTERNATIVE (MLT Capacity Volumes) 48 EXHIBIT 4-13 INTERSECTION PERFORMANCE MEASURES FOR THE PARKWAY- MINOR ARTERIAL (MILE 7) INTERSECTION WITH MLT CAPACITY VOLUMES 49 EXHIBIT 4-14 PARKWAY-MINOR ARTERIAL INTERSECTION (MILE 7) DELAY PER MOVEMENT BY INTERSECTION DESIGN ALTERNATIVE (MLT Capacity Volumes) 50 EXHIBIT 4-15 INTERSECTION PERFORMANCE MEASURES FOR THE PARKWAY- PARKWAY (MILE 6) INTERSECTION WITH MLT UNIFORM CAPACITY VOLUMES 51 EXHIBIT 4-16 PARKWAY-PARKWAY INTERSECTION (MILE 6) DELAY PER MOVEMENT BY INTERSECTION DESIGN ALTERNATIVE (MLT Uniform Capacity Volumes) 53 EXHIBIT 4-17 COMPARISON OF DELAY BY MOVEMENT FOR THE MLT AND CFI DESIGNS WITH CAPACITY VOLUMES 56 EXHIBIT 4-18 CFI CAPACITY TRAFFIC OPERATIONS 58 EXHIBIT 4-19 CFI DELAY BY MOVEMENT FOR CFI CAPACITY AND UNIFORM CAPACITY VOLUMES 59 Phase 2 Continuous Flow Intersections

9 LIST OF EXHIBITS Page EXHIBIT 4-20 EXHIBIT 4-21 EXHIBIT 4-22 EXHIBIT 4-23 CFI AND SPUI TRAFFIC OPERATIONS WITH CFI UNIFORM CAPACITY TRAFFIC VOLUMES 60 DELAY PER VEHICLE BY MOVEMENT FOR THE CFI AND SPUI WITH CFI UNIFORM CAPACITY VOLUMES 61 MCDOT TYPICAL SECTION FOR AN URBAN PRINCIPAL ARTERIAL ROADWAY 63 AERIAL VIEW OF A CFI WITH FRONTAGE ROADS FOR ACCESS TO ADJACENT PROPERTIES 64 EXHIBIT 5-1 COMPARISON OF CFI TO MLT DESIGN BY ANALYSIS SCENARIO 67 EXHIBIT 5-2 COMPARISON OF SPUI TO CFI DESIGN BY ANALYSIS SCENARIO 67 Phase 2 Continuous Flow Intersections

10

11 EXECUTIVE SUMMARY STUDY OVERVIEW Long range transportation studies in the Phoenix metropolitan area have identified the need for non-freeway restricted access facilities capable of providing significantly greater travel capacity than that provided by major urban arterials. Various design and access refinement alternatives have been proposed to provide additional travel capacity without employing full gradeseparations at intersections with arterial cross-streets. These design alternatives can provide the benefit of increases in intersection capacity while maintaining the potential for direct driveway access to each quadrant of the intersection. These design alternatives have also demonstrated significant safety benefits over standard intersection designs. These innovative design alternatives generally focus on the provision of simple two-phase traffic signal operations at the intersections by eliminating the left-turn movement at the intersection and accommodating it elsewhere. One of these alternatives is referred to as a continuous flow intersection (CFI), although this is a misnomer in that traffic is required to stop for traffic signals at the intersection. The general CFI configuration is illustrated in Exhibit ES-1. The basic concept of the CFI is to separate the left-turns from through movements upstream of the main intersection, and store these movements to the left of the opposing through movements. At the main intersection, the left-turns are made at the same time as the opposing through movement. This provides for simple two-phase operation of the main intersection when the left-turns are removed on all four intersection approaches. This design scheme also provides for free-flow right-turn movements on each approach that do not enter the main intersection, further reducing intersection traffic delay and congestion. A prior study was conducted to evaluate an indirect or Michigan Left-Turn (MLT) concept along a hypothetical eight-lane parkway corridor and compare that against a standard restricted access eight-lane arterial configuration composed of conventional at-grade, eight-phase signalized intersections 1. Traffic operations, traffic delay, and general roadway capacity measures were used to assess the differences between the base case and the MLT. This prior study also analyzed the performance of a grade-separated single point urban interchange (SPUI) versus an all-direction MLT design. The purpose of this study is to expand on the prior work by evaluating and comparing the traffic operations of CFIs to the performance of standard intersections, MLT intersections, and the SPUI design under similar conditions and assumptions as used previously. STUDY METHODOLOGY The approach to this study was similar to that of the prior work, but with some differences to accommodate limitations of the software used. That is, the same hypothetical 12-mile long parkway corridor used in the prior study was also used here, except that not all intersections were converted to CFI intersections. In this study only three intersections along the corridor were evaluated as CFIs. These three intersections consisted of a major parkway-parkway intersection, a parkway-major arterial intersection, and a parkway-minor arterial intersection. This approach allowed the comparison of aggregate metrics as well as comparison of traffic operations at individual intersections and for individual traffic movements. 1 MCDOT Enhanced Parkway Study, Final Report, Maricopa County Department of Transportation, Contract No , August Phase 2 Continuous Flow Intersections ES-1

12 Exhibit ES-1 ILLUSTRATION OF A CONTINUOUS FLOW INTERSECTION Source: ABMB Engineers, Inc., Baton Rouge, LA. Reproduced with permission. The simulated corridor consisted of two density variants; a high intensity urban condition, and a lower intensity suburban condition. Due to limitations in the software used relating to intersection spacing, the CFIs were only simulated along the lower intensity suburban section of the corridor. The high intensity urban condition consists of the following characteristics: The parkway is represented as an eight-lane divided roadway between intersections, with intersection turn lanes commensurate with the type of intersecting roadway. Major arterial intersections every mile with traffic signal control. Minor arterial intersections at half-mile spacing between the major arterials with traffic signal control. Stop controlled collector roadway intersections at quarter-mile spacing between the major arterials and the minor arterials. The division between the high-intensity urban and suburban conditions (Mile 6 roadway) is represented by another eight-lane divided parkway. Phase 2 Continuous Flow Intersections ES-2

13 The lower intensity suburban condition consists of the following characteristics: The parkway is represented as an eight-lane divided roadway between intersections, with intersection turn lanes commensurate with the type of intersecting roadway. Major arterial intersections every two miles with traffic signal control. Minor arterial intersections at one mile spacing between the major arterials with traffic signal control. Stop controlled collector intersections at ½-mile spacing between arterial intersections. The traffic simulation was conducted using Synchro/SimTraffic, a traffic operations analysis software package developed by TrafficWare. The following general network scenarios represented the roadway configurations used to make the comparisons: Base Network A network consisting of conventional at-grade signalized and stopcontrolled intersections consistent with the Maricopa County intersection design standards for each intersection type. The intersections provided exclusive left-turn and right-turn lanes consistent with the traffic demand, and protected left-turn movement signal timing. MLT Network All of the intersections along the main parkway were converted to MLT intersections. A typical schematic representation of an MLT intersection is provided in Exhibit ES-2. U-turn opportunities were provided on the far side of the parkway intersection to accommodate the left-turn movements. At the Mile 6 parkway-parkway intersection, U-turn opportunities were provided on all four legs of the intersection. A median of a minimum 60 feet in width was assumed as part of the main parkway. Such a median is typical for the provision of U-turn opportunities, and is required to provide adequate traffic operations for the U-turn movement. MLT Network with a Single Point Urban Interchange (SPUI) at the parkway-parkway intersection The parkway-parkway intersection was simulated with a grade separated SPUI, while the remainder of the intersections were MLT designs. CFI Application The CFIs were applied at the same three intersections in the Base Network and in the MLT network. These intersections were: - Mile 6 parkway-parkway intersection. - Mile 7 parkway-minor arterial intersection. - Mile 10 parkway-major arterial intersection. The traffic flow simulation study enabled the evaluation of traffic operations, including delay, travel time, and number of stops on the three network alternatives. In addition, tests were constructed to estimate the capacity of the Base and MLT Networks, and the capacity of the CFI intersections. Phase 2 Continuous Flow Intersections ES-3

14 Exhibit ES-2 EXAMPLE OF A TYPICAL MLT INTERSECTION Source: Source: SUMMARY OF RESULTS AND CONCLUSIONS The following results and conclusions are based on the technical analysis conducted in this study. It should be noted that the results and conclusions based on the traffic simulation conducted for this study may be limited to the range of conditions applied and assumptions made in conducting the analysis, and cannot necessarily be generalized to all possible combinations of conditions. The capacity of an eight-lane divided roadway consisting of CFI intersections is estimated to be between 92,000 and 108,000 vehicles per day (vpd) assuming that the peak-hour volume is between 10 and 8.5 percent of the daily traffic. This result is considered somewhat conservative in that the capacities of the left-turn and right-turn lanes were not explored in this study, and these lanes operated under capacity with the assumed traffic volumes. Considering only the through movement at signalized intersections, the CFI design capacity is estimated to be approximately 1,000 to 1,050 vehicles per hour per through lane, which is slightly higher than the estimated 975 to 1,025 vphpl for the MLT design. The primary difference in through movement capacity of the CFI and the MLT intersection is that with the MLT intersection the left-turn movement volume must pass through the intersection, reducing the capacity for through traffic. The capacity of a CFI is estimated to be between 45 and 55 percent greater than the capacity of a conventional intersection design with multi-phase signal timing. Phase 2 Continuous Flow Intersections ES-4

15 The aggregate intersection performance measures indicate that the CFI reduced delay by 46 percent, reduced stops by 50 percent, and reduced travel time by 46 percent in comparison to the conventional intersection design. At the parkway-parkway (Mile 6) intersection the CFI reduced delay by 59 percent, stops by 69 percent, and travel time by 59 percent in comparison to the conventional intersection design. The CFI is similar to the MLT in terms of the treatment of the through traffic movement at the intersection. Both intersection designs provide simple two-phase signal timing for the through movement, significantly increasing through movement capacity in comparison to the multi-phase signal timing used with conventional intersection design. The primary reduction in delay, stops, and travel time with the CFI design is in the leftturn and right-turn movements in comparison to the MLT design. In the CFI, the leftturns only pass through the intersection configuration once as opposed to the MLT design where they pass through the intersection twice. The free flow right-turn lanes also provide significant advantage to the CFI design. A summary comparison of the performance metrics for the MLT intersection design and the CFI is provided in Exhibit ES-3 for each analysis scenario. These results indicate that the full CFI (left-turn traffic removed upstream of the intersection on each approach) provided significant reductions in delay, stops, and travel time in comparison to the MLT across all analysis scenarios and for each of the intersection types evaluated. The minimum reduction in delay for a full CFI in comparison to the MLT design was 19 percent, with a maximum of 55 percent. Free flow right-turn lanes could be included in the MLT design on each approach that provides the median U-turn movement for left-turns, but this feature was not tested as part of this study. The use of free flow right-turn lanes with the MLT design would reduce delay to both the right-turn movement and the associated left-turn movement, and could significantly improve the traffic operations of the MLT design. The CFI generated better performance metrics than the SPUI for the Base Capacity volumes (see Exhibit ES-4), which were the lowest volumes tested. The SPUI generated significantly better performance metrics than the CFI with the MLT Capacity and CFI Uniform Capacity volumes. The SPUI and the CFI provided comparable levels of delay with the MLT Uniform Capacity Volumes. The traffic operations performance improvements for the SPUI were as high as a 58 percent reduction in delay, a 69 percent reduction in stops, and a 39 percent reduction in travel time in comparison to the CFI design. The primary benefit of the SPUI design is the free flow operation of the grade separation for the major through movement of traffic on the parkway in the analysis. While it may be possible for the CFI to provide traffic operations characteristics similar to the SPUI at lower volume levels, this was not the case for the higher traffic volumes that approximated the capacity of the CFI intersection used in this study. A potential advantage of the CFI design over the SPUI is the ability of the CFI to accommodate high volumes of left-turn and right-turn movements with lower levels of delay than the SPUI. Phase 2 Continuous Flow Intersections ES-5

16 A disadvantage of the CFI design is the issue of providing access to adjacent properties on the corners of the intersection. The location of the left-turn lanes and the free flow right-turn lanes basically eliminates access within the limits of these lanes. Direct access should not be provided from the free flow right-turn lanes because of safety concerns. An alternative that has been used is to provide access to properties at the intersection via a frontage road system that connects to the main roadway down stream of the end of the free flow right-turn lane. Placement of the access point and design of the frontage road system would very much depend on the individual development plan for the adjacent property. These access points should be right-in, right-out only. The minimum right-of-way for an eight-lane divided parkway with CFI intersections consisting of dual left-turn lanes and a single free flow right-turn lane on each approach would be approximately 201 feet. This ROW is consistent with the estimated requirements for an MLT intersection corridor of approximately 200 feet, and is 24 feet less than the estimated 225 feet required for an MLT design at the parkway-parkway intersection. Exhibit ES-3 COMPARISON OF CFI TO MLT DESIGN BY ANALYSIS SCENARIO Volume Scenario Base Capacity MLT Capacity MLT Uniform Performance Metric CFI Percent Difference Analysis Travel Location Delay Stops Time Aggregate for 3 Intersections Parkway-Parkway Intersection (Mile 6) Parkway-Major Arterial Intersection (Mile 10) Parkway-Minor Arterial 1 Intersection (Mile 7) Aggregate for 3 Intersections Parkway-Parkway Intersection (Mile 6) Parkway-Major Arterial Intersection (Mile 10) Parkway-Minor Arterial Intersection (Mile 7) Parkway-Parkway Intersection (Mile 6) Capacity 1. Only a partial CFI was simulated at this location for this volume scenario, that is, the left-turns were removed from the main intersection for only the two approaches on the main parkway. Phase 2 Continuous Flow Intersections ES-6

17 Exhibit ES-4 COMPARISON OF SPUI TO CFI DESIGN BY ANALYSIS SCENARIOS Volume Scenario Base Capacity MLT Capacity MLT Uniform Capacity CFI Uniform Capacity Performance Metric SPUI Percent Difference Analysis Travel Location Delay Stops Time Parkway-Parkway Intersection (Mile 6) Parkway-Parkway Intersection (Mile 6) Parkway-Parkway Intersection (Mile 6) Parkway-Parkway Intersection (Mile 6) Phase 2 Continuous Flow Intersections ES-7

18

19 1. INTRODUCTION AND STUDY OVERVIEW Long range transportation studies in the Phoenix metropolitan area have identified the need for non-freeway restricted access facilities able to offer significantly greater travel capacity than that provided by major urban arterials. Various design and access refinement alternatives have been proposed to provide additional travel capacity without employing full grade-separations at intersections with arterial cross-streets. These design alternatives provide the benefit of increases in intersection capacity while maintaining the potential for direct driveway access for commercial or other use to each quadrant of the intersection. These design alternatives may also provide significant safety benefits over standard intersection designs. These innovative design alternatives generally focus on the provision of simple two-phase traffic signal operations at the intersections by eliminating the left-turn movement at the intersection and accommodating it elsewhere. One of these alternatives is referred to as a continuous flow intersection (CFI), although this is a misnomer in that traffic is required to stop for traffic signals at the intersection. The general CFI configuration is illustrated in Exhibit 1-1. The basic concept of the CFI is to separate the left-turns from through movements upstream of the main intersection, and store these movements to the left of the opposing through movements. At the main intersection, the left-turns are made at the same time as the opposing through movement. This provides for simple two-phase operation of the main intersection when the left-turns are removed on all four intersection approaches. This design scheme also provides for free-flow right-turn movements on each approach that do not enter the main intersection, further reducing intersection traffic delay and congestion. A prior study was conducted to evaluate an indirect or Michigan Left-Turn (MLT) concept along a hypothetical eight-lane parkway corridor and compare that against a standard restricted access eight-lane arterial configuration composed of conventional at-grade, eight-phase signalized intersections 2. Traffic operations, traffic delay, and general roadway capacity measures were used to assess the differences between the base case and the MLT. This prior study also analyzed the performance of a grade-separated single point urban interchange (SPUI) versus an all-direction MLT design. A typical MLT intersection design is provided in Exhibit 1-2 for reference. This study is a companion work to the prior study and expands on that analysis by including the CFI in the comparison of alternatives. The purpose of this study is to evaluate and compare the traffic operations characteristics of CFIs to the performance of standard intersections, MLT intersections, and the SPUI design under similar conditions and assumptions as used previously. 2 MCDOT Enhanced Parkway Study, Final Report, Maricopa County Department of Transportation, Contract No , August Phase 2 Continuous Flow Intersections 1

20 Exhibit 1-1 EXAMPLE OF A CONTINUOUS FLOW INTERSECTION Source: ABMB Engineers, Inc., Baton Rouge, LA. Reproduced with permission. Phase 2 Continuous Flow Intersections 2

21 Exhibit 1-2 EXAMPLE OF A TYPICAL MLT INTERSECTION Source: Source: Phase 2 Continuous Flow Intersections 3

22

23 2. GENERAL ANALYSIS APPROACH AND METHODOLOGY BASE NETWORK CONFIGURATION The general analysis approach for this study was to create a hypothetical 12-mile long parkway corridor consisting of two six-mile variants. One case represents a high intensity development urban location, and the other represents a somewhat lower intensity suburban area. The high intensity urban condition consists of the following characteristics: The parkway is represented as an eight-lane divided roadway between intersections, with intersection turn lanes commensurate with the type of intersecting roadway. Major arterial intersections every mile with traffic signal control. Minor arterial intersections at half-mile spacing between the major arterials with traffic signal control. Stop controlled collector or commercial access roadway/driveway intersections at quarter-mile spacing between the major arterials and the minor arterials. The division between the high-intensity urban and suburban conditions (Mile 6 roadway) is represented by another eight-lane divided parkway. The lower intensity suburban condition consists of the following characteristics: The parkway is represented as an eight-lane divided roadway between intersections, with intersection turn lanes commensurate with the type of intersecting roadway. Major arterial intersections every two miles with traffic signal control. Minor arterial intersections at one mile spacing between the major arterials with traffic signal control. Stop controlled collector or commercial access driveway intersections at ½-mile spacing between arterial intersections. The general corridor roadway condition assumed for the analysis is schematically illustrated in Exhibit 2-1. The location of the CFI applications is also provided in Exhibit 2-1. The CFIs were applied to the Base Network and the MLT Network. A single SPUI intersection design was also applied to corridor at the parkway-parkway intersection illustrated in Exhibit 2-1. For discussion purposes and presentation of results, the main parkway corridor is assumed to run north-south, with cross streets east-west. Major arterials are assumed to consist of a sixlane roadway. Minor arterials are assumed to consist of a four-lane roadway, and collectors are assumed to consist of a two-lane roadway. The parkway and all cross streets are assumed to provide exclusive turn lanes consistent with the type of roadway being evaluated. The intersection number of lanes and lane use assumed for each type of intersection included in the Base Network are provided in Exhibit 2-2. MLT NETWORK CONFIGURATION The MLT Network configuration was established to provide a direct comparison to the Base Network. Hence, the parkway and the cross streets were all assumed to have the same number of through lanes as in the Base Network. Left-turn lanes at the parkway intersections Phase 2 Continuous Flow Intersections 4

24 were eliminated. The number of lanes and lane use assumptions for the initial MLT Network condition are provided in Exhibit 2-3. The MLT Network configuration consists of the following conditions: The parkways are eight-lane divided facilities, offset by 120-feet between the centers of the arterial roadways. This provides a median of approximately 70 feet in width for the simulation. The parkway-parkway intersection was assumed to have a directional U-turn on each intersection approach. For consistency, parkway traffic was assumed to travel through the main intersection and then U-turn and right-turn to accomplish the left-turn movement. A grade separated single point urban intersection (SPUI) was selected as a design alternative to be tested for the parkway-parkway intersection. The through traffic movement on the SPUI ramps at the cross street was not allowed in order to provide the most efficient SPUI traffic operations. Dual right-turn lanes were assumed on all approaches of the parkway-parkway and parkway-arterial intersections to accommodate the high left-turn and right-turn volumes assumed in the Base Case. Single right-turn lanes were assumed all parkway-minor arterial and parkway-collector intersection approaches. Directional U-turn opportunities were placed in the parkway median adjacent to the cross street intersections. The following characteristics were assumed for the directional U- turn location, number of turn lanes, and traffic control for each cross street type: feet from major arterials, dual turn lanes, traffic signal controlled feet from minor arterials, single U-turn lanes, signal controlled. This condition was previously simulated with dual U-turn lanes with a stop control, but questions were raised regarding the functionality of this configuration from a safety perspective. Therefore, the configuration was changed for this analysis. This modification of the original configuration of the previous study had no impact on the MLT Network system-wide results feet from collectors, single turn lane, stop controlled. - Link distances are measured from the center of the intersection. Phase 2 Continuous Flow Intersections 5

25 Exhibit 2-1 GENERAL ROADWAY SCHEME FOR THE ANALYSIS LEGEND Maj Major Arterial Min Minor Arterial Q Quarter Mile Spacing C Collector/ Business Access Signalized Entry Intersection Signalized Intersection Stop-Controlled Intersection SPUI Application CFI Application Mile 12 Maj Half 11 C Mile 11 Min Half 10 C Mile 10 Maj Half 9 C Mile 9 Min N Note Signalized entry intersections are placed one-half mile upstream of the intersection they are metering. Half 8 C Half 7 C Mile 7 Min Mile 8 Maj Suburban Area High Intensity Area Half 6 C Mile 6 Parkway Q 5b C Half 5 Min Q 5a C Mile 5 Maj Q 4b C Half 4 Min Q 4a C Q 3b C Q 3a C Half 3 Min Mile 4 Maj Mile 3 Maj Q 2b C Half 2 Min Q 2a C Mile 2 Maj Q 1b C Half 1 Min Q 1a C Mile 1 Maj Phase 2 Continuous Flow Intersections 6

26 Exhibit 2-2 BASE NETWORK INTERSECTION LANE CONFIGURATION, LANE USE, AND INTERSECTION CONTROL ASSUMPTIONS Parkway / Parkway Intersection Parkway / Major Arterial Intersection Parkway / Minor Arterial Intersection Parkway / Collector or Commercial access Intersection LEGEND Number of Lanes and Lane Usage Signalized Intersection Stop-Controlled Intersection Phase 2 Continuous Flow Intersections 7

27 Exhibit 2-3 MLT NETWORK LANE CONFIGURATION, LANE USE AND INTERSECTION CONTROL ASSUMPTIONS Parkway / Parkway Intersection Parkway / Major Arterial Intersection Parkway / Minor Arterial Intersection Parkway / Collector or Commercial access Intersection LEGEND Number of Lanes and Lane Usage Signalized Intersection Stop-Controlled Intersection Phase 2 Continuous Flow Intersections 8

28 SINGLE POINT URBAN INTERCHANGE (SPUI) CONFIGURATION A schematic diagram of the assumed SPUI configuration is provided in Exhibit 2-4. The SPUI configuration was assumed such that the through movement on the main parkway is grade separated and does not pass through the signalized intersection. Only three through lanes in each direction are carried through the intersection as one lane in each direction becomes an exit ramp. Dual left-turn lanes are provided on each approach, with a single right-turn lane. There is no through movement from the ramp across the cross street, which provides for the most efficient operation of a SPUI as this eliminates a signal phase, providing a simplified threephase signal operation. Overall, this configuration provides a relatively compact, efficient intersection design. Exhibit 2-4 ASSUMED SPUI CONFIGURATION FOR THE PARKWAY-PARKWAY INTERSECTION Free flow through movements Free flow through movements LEGEND Number of Lanes and Lane Usage Signalized Intersection Phase 2 Continuous Flow Intersections 9

29 CONTINUOUS FLOW INTERSECTION CONFIGURATION A schematic diagram of the CFI number of lanes and lane configuration for each intersection type is provided in Exhibit 2-5. Only a partial CFI was used for the application on the Base Network at the minor arterial intersection. That is, only the approaches on the parkway included the CFI treatment for right and left-turn movements at the minor arterial intersection. This configuration worked very well with the Base capacity traffic volumes, but did not function very well with the MLT capacity volumes, and therefore a full CFI was used at the minor arterial intersection with the MLT and CFI capacity volume scenarios. The left-turns at a CFI cross the opposing through lanes at a point approximately 500 feet upstream of the main intersection. For the purposes of this analysis, the distance from the leftturn location to the main intersection was established to provide efficient traffic operations for the left-turn movement by minimizing the additional delay to this movement. With proper placement of the left-turn location relative to the main intersection, the left-turn signal timing and offsets could be established such that good signal coordination was provided for the left-turns without increasing the traffic delay to the through movements in the intersection. The right-turn movements at a CFI are free flow and do not pass through any of the traffic signals. For the purposes of this analysis, acceleration lanes of sufficient length were provided for right-turn traffic to allow for proper merging with minimal delay to this movement. Phase 2 Continuous Flow Intersections 10

30 Exhibit 2-5 CFI INTERSECTION SCHEMATIC CONFIGURATION AND TRAFFIC CONTROL Parkway / Parkway Intersection Parkway / Major Arterial Intersection Parkway / Minor Arterial Intersection LEGEND Number of Lanes and Lane Usage Signalized Intersection Phase 2 Continuous Flow Intersections 11

31 TRAFFIC ANALYSIS METHODOLOGY The analysis of traffic operations was conducted using the Synchro/SimTraffic software package developed by TrafficWare, Corporation. Synchro is a deterministic traffic operations, and traffic signal timing optimization software package. Synchro implements the methods of the 2000 Highway Capacity Manual for urban streets, signalized intersections, and unsignalized intersections. In addition to calculating capacity and level of service, Synchro can also optimize traffic signal cycle lengths, splits, and offsets for intersections along a corridor. SimTraffic is companion micro-simulation software to Synchro. SimTraffic uses the Synchro data entry and the traffic signal timing generated by Synchro to create a stochastic microsimulation of the entire roadway network under consideration. Each vehicle entering the network is tracked individually, and the traffic operations metrics are generated by compiling the individual vehicle information along each segment of the roadway. SimTraffic also creates an animation of the network showing each vehicle s journey and interaction with other vehicles. For this analysis Synchro was used for creating the network to be evaluated, data entry, and traffic signal optimization. Due to the complex nature of the traffic movements through the MLT network and the CFIs, SimTraffic was used to evaluate the traffic operations metrics and report the system performance of the alternatives. The simulation of each condition was executed five times using different random number seeds for each execution. The results of the five simulations were compiled and averaged to represent the final results for comparison of the alternatives being evaluated. SimTraffic assumes random arrivals of vehicles on the first roadway segment entering a network. From that point on, vehicle arrivals at intersections are dictated by the traffic signal timing of upstream signals and the vehicle travel speeds on the network. Typically, random arrivals of vehicles at an intersection will result in higher estimates of vehicle delay than with a condition where arrivals are metered by upstream signals in a coordinated signal system. Therefore, in order to reduce the effects of random intersection arrivals on the analysis, entry intersections were used on each of the parkways and major arterials. The entry intersections are the first intersections encountered by traffic on these roadways in the simulation, thus eliminating the random arrivals at the next downstream intersection. Entry intersections were assumed to be one-half mile upstream of the parkway intersections. The locations of the entry intersections are also provided in Exhibit 2-1. The delay and traffic operations conditions at these entry intersections were not included in the performance metrics for the analysis of alternatives. The parkway corridor in the MLT condition was simulated as two parallel one-way streets. This allows for proper simulation of the presence of the median and the directional median U-turns. In addition, it provides better simulation of the major signalized intersections. When the CFIs were added to the MLT network, the parallel one-way street configuration was maintained at the CFI. This creates a slightly longer travel path (60 to 120 feet longer) for the left-turns through the intersection at a CFI, but it was not considered significant in the overall results. In a real application of a CFI along an MLT parkway, the median would most likely be reduced in size at the intersection. TRAFFIC VOLUMES Traffic volumes at capacity for the Base Network, MLT Network, and the MLT Network with CFIs were established to represent peak-hour LOS E (capacity) conditions along the simulated Phase 2 Continuous Flow Intersections 12

32 corridor at each intersection. The idea was to approximate the limits of the corridor design to accommodate traffic demand. The general process used to determine the capacity volumes is described in the final report for the previous study 3. The Base Capacity volumes for the Base Network were used as the initial volumes for the MLT Network for the comparison of traffic operations. The capacity volumes for the MLT Network were used as the initial volumes for the MLT Network with the CFIs. Note that for the MLT Capacity and the CFI Capacity volumes, only the main parkway through volumes were increased, while cross street volumes and turn movement volumes were held constant. This was done to provide an estimate of the through volume capacity for the MLT and CFI intersections. An additional set of traffic volumes was developed for use in testing the traffic operations at capacity for the parkway-parkway intersection with the MLT, SPUI and CFI design. This set of traffic volumes provides approximately uniform demand on each intersection approach. SIMULATION OF MLT AND CONTINUOUS FLOW INTERSECTIONS SimTraffic is a sophisticated and powerful traffic simulation tool that is capable of precisely recreating most intersection designs and traffic signal control concepts in the simulation environment. In the previous study, SimTraffic was found capable of simulating an MLT intersection with more than sufficient accuracy to provide a valid comparison of traffic operating characteristics. The processes used to simulate MLT traffic operations and compile traffic statistics are described in detail in the previous study. An animated image of a MLT intersection from the simulation is provided in Exhibit 2-6 for the parkway-major arterial intersection. While the CFI operation appears complex, it is relatively straight forward from a simulation perspective. The traffic movements at each node within the overall intersection are designated in a straight forward manner without the requirement of assigning movement volumes in node pairs as was the case for the MLT intersection. The primary difficulty was creating the nodes for the left-turn lane intersections with the cross street while maintaining the minimum node spacing required by the software. As mentioned earlier, care was also taken to position the left-turn nodes far enough upstream from the main intersection to provide the spacing necessary to allow for proper signal timing to coordinate the left-turn movement signals while maintaining proper signal coordination for the through movements at the intersection. Care was also taken to provide sufficient signal clearance intervals to insure that cross street through traffic cleared the left-turn lane intersections before releasing the left-turn movement. An animated image of the simulation of a CFI intersection at the parkway-major arterial intersection on the Base Network is provided in Exhibit 2-7. This image is a reasonable approximation of how an actual CFI might look in terms of size and the relative location of the lanes within the intersection. Exhibit 2-8 provides an animated image of the simulation of a CFI at the parkway-major arterial intersection on the MLT Network. The MLT network was simulated as two parallel one-way streets to accurately represent the presence of the median and the U-turn movement through the directional median openings. This image may not be a particularly good representation of a 3 MCDOT Enhanced Parkway Study, Final Report, Maricopa County Department of Transportation, Contract No , August Phase 2 Continuous Flow Intersections 13

33 CFI on an MLT parkway in that it is likely that the median would be reduced in size at the intersection to save right-of-way, and the intersection would look like that represented in Exhibit 2-7. However, from a traffic operations perspective, the CFI intersections on the MLT network worked very well and provided for a good comparison of alternatives. It should be noted that the simulation of the CFI on the MLT network required the use of slightly longer clearance intervals for through traffic to clear the left-turn intersections. Therefore, under CFI capacity conditions, the capacity results can be considered somewhat conservative. It should also be noted that the delay to through and left-turning vehicles resulting from travel through the median at the main intersection in the simulation was not included in the summary of results for the CFI on the MLT network. This was done considering that the CFI would not be designed as it appears in Exhibit 2-8, but would look like the design in Exhibit 2-7. Exhibit 2-6 SIMULATION IMAGE OF AN MLT INTERSECTION Phase 2 Continuous Flow Intersections 14

34 Exhibit 2-7 SIMULATION IMAGE OF A CFI ON THE BASE NETWORK Phase 2 Continuous Flow Intersections 15

35 Exhibit 2-8 SIMULATION IMAGE OF A CFI ON THE MLT NETWORK Phase 2 Continuous Flow Intersections 16

36 ALTERNATIVE ANALYSIS SCENARIOS Five alternative analysis scenarios were developed for this study to test various traffic operations characteristics of the corridor. The following provides a description of these scenarios and how they were used by the study: 1. Base Network with Base Capacity Volumes: The Base Network is the network with the conventional intersection and corridor configuration described above. The traffic volumes were adjusted to provide traffic operations at or near capacity (LOS E, 55 to 80 seconds of delay per vehicle by approach) as defined by the 2000 Highway Capacity Manual for as many of the intersection approaches as practical while maintaining balanced intersection volumes between intersections. This scenario was used to define the approximate maximum peak-hour traffic volume capability of the corridor with the Base Network. The Base Capacity volumes were used to evaluate the following alternative network configurations: Base Network (previous study results used). Base Network with CFIs at Mile 6 (parkway-parkway intersection), Mile 7 (parkwayminor arterial intersection), and Mile 10 (parkway-major arterial intersection). MLT Network (previous study results were updated by replacing eight stop-controlled dual U-turn lane locations with signal controlled single U-turn lanes). MLT Network with a SPUI at the Mile 6 parkway-parkway intersection (previous study results were updated using improved signal timing scheme). The entire 12-mile corridor was used in the simulation for each alternative network. The manner in which the performance metrics were compiled allowed the comparison of aggregate intersection performance, and allowed for the comparison of individual intersections and traffic movements. 2. MLT Network with MLT Capacity Volumes: The through traffic volumes for the main parkway corridor were increased until the parkway approach delay per vehicle represented LOS E traffic operations for as many of the intersections as practical. The approach volumes on the cross streets were those volumes used in the Base Capacity scenario. This scenario was used to approximate the upper limit of peak-hour capacity for the MLT parkway, while providing approximately LOS E operation for the cross streets. That is, parkway traffic volumes were increased, and signal timing adjusted to favor the parkway traffic, until near capacity operations were exhibited along the corridor. The following network simulations were conducted for comparison using the MLT Capacity volumes: MLT Network (previous study results were updated by replacing two stop-controlled dual U-turn lane locations with signal controlled single U-turn lanes). MLT Network with SPUI at the Mile 6 parkway-parkway intersection (previous study results used). MLT Network with CFIs at Mile 6 (parkway-parkway intersection), Mile 7 (parkwayminor arterial intersection), and Mile 10 (parkway-major arterial intersection). The entire 12-mile corridor was used in the simulation for each alternative network. The manner in which the performance metrics were compiled allowed the comparison of aggregate intersection performance, and allowed for the comparison the performance of individual intersections and traffic movements. Phase 2 Continuous Flow Intersections 17

37 3. CFI Network Capacity Volumes: This scenario increased the northbound and southbound through traffic volumes entering the major arterial and minor arterial CFIs on the MLT Network. This was done to estimate the network capacity of the CFIs and compare this to the network capacity for the MLT intersections developed in the previous study. Traffic volumes at the parkway-parkway intersection were not increased because this intersection was judged to be operating at capacity with the MLT Capacity volumes. 4. Parkway-Parkway Intersection with Uniform MLT Capacity Volumes: This scenario focused on only the parkway-parkway intersection capacity. The simulated roadway network was reduced to only the parkway-parkway intersection and the adjacent signalized intersections upstream and downstream of the parkway-parkway intersection. The traffic volumes on each MLT parkway were adjusted to provide relatively equal volumes on all four approaches to estimate the approach capacity when the available green time was split equally between the two parkways. The following intersection alternatives were simulated: MLT intersection single U-Turn lanes on each intersection leg 4 (previous study results used). SPUI intersection with dual left-turn lanes on each approach (previous study results used). CFI intersection with dual left-turn lanes on each approach. 5. Parkway-Parkway Intersection with Uniform CFI Capacity Volumes: The purpose of this scenario was to attempt to estimate capacity volumes for the CFI assuming relatively uniform demand on each approach. The traffic volumes from the MLT Uniform Capacity condition were increased until an approximate LOS E (capacity) condition was generated for each approach of the CFI. The following intersection alternatives were evaluated using the CFI uniform capacity volumes: SPUI intersection with dual left-turn lanes on each approach. CFI intersection with dual left-turn lanes on each approach. TRAFFIC SIGNAL TIMING FOR THE ANALYSIS AND COMPARISON OF ALTERNATIVES As a method of providing a non-bias assessment and comparison of alternatives, Synchro was used to optimize traffic signal parameters for the simulations. Traffic signal cycle length, intersection splits, and offsets were optimized. The intersections along the main parkway corridor were assigned to the same zone for optimization of the cycle lengths and offsets. This provided the best condition for maintaining traffic progression along the main parkway as the entry intersections were not included in establishing the cycle lengths and offsets along the parkway. All of the intersections along the main parkway, including the MLT signalized U-turn locations and the continuous flow intersections, operated on the same cycle length, except for the SPUI, which was optimized separately from the remainder of the parkway. In general, Synchro s signal timing algorithms tend to favor the highest traffic demand roadway, in this case the parkway, in establishing signal timing. The systemwide metrics used by Synchro may produce signal timings resulting in LOS F conditions on the cross street, if this reduces overall delay. The animation generated by the simulation and Synchro/SimTraffic 4 The analyses conducted in the previous study indicated that the use of dual U-turn lanes or single U-turn lanes produced essentially the same results for the left-turn volumes tested at the Mile 6 intersection. Only the results for the single U-turn lane condition were included in this study. Phase 2 Continuous Flow Intersections 18

38 congestion parameters were reviewed to determine whether the optimized signal timings were resulting in what was considered unreasonable cross street congestion (for example, LOS F with over 150 seconds of delay per vehicle). In such cases, the signal timings were adjusted to provide more efficient cross street traffic operations. These and other efforts were made to provide the most efficient and realistic traffic operation possible with the demand volumes used in the analysis. For the Base Network, all left-turn phases were protected only and were assigned as lead leftturns. SimTraffic default parameters for minimum headway, turn speed, and start-up delay were set to maximize left-turn lane throughput to provide the most efficient Base Case traffic operations possible. Minimum green times for pedestrians were not considered in the analysis, and pedestrian traffic was not included in the simulation of alternatives. Traffic signal change plus clearance interval time was estimated for each intersection type based on the size of the cross street and the assumed intersection approach speeds for each roadway type. The equation for calculating the signal change plus clearance interval time presented in the Institute of Transportation Engineers, Traffic Engineering Handbook 5 was used for these calculations. The resulting times were rounded to the nearest half second for use in the traffic analysis. Signal timing at signalized MLT U-turn locations was coordinated with the downstream traffic signal. Offsets were optimized to provide good progression and to allow efficient movement of the U-turn. Placement of the upstream left-turn locations for the continuous flow intersections and signal timing for the CFI left-turns was established to achieve good progression for the left-turn movement through the downstream left-turn signals. The signal timing also provided signal coordination and progression for the main parkway through movements to the extent possible under the various traffic demand scenarios. PERFORMANCE ANALYSIS METRICS The primary performance metrics for this analysis are total delay (vehicle hours), delay per vehicle (seconds per vehicle), total stops, stops per vehicle, and total travel time in the system as measured and compiled by the SimTraffic program. These primary metrics were summarized for the system as a whole and for individual intersections by intersection approach. The system consists of all roadway elements in the simulation excluding the entry intersections and entry intersection approaches. The total number of vehicles entering the network was used as a comparison between alternatives to ensure that the performance metrics were developed based on the same level of vehicular activity on the network. The following definitions are provided for clarity: Delay SimTraffic calculates delay as the difference between the simulated travel time over a defined link distance and the time it would take the vehicle with no other vehicles or traffic control devices present. Total delay is the sum of the delay for each vehicle in the simulation. Delay per vehicle is the total delay divided by the total number of vehicles. 5 Institute of Transportation Engineers, Traffic Engineering Handbook, 5 th Edition, 1999, page 481. Phase 2 Continuous Flow Intersections 19

39 Simulation volume is the number of vehicles counted as in the system for the simulation. This number will differ slightly for each random simulation, even when the same input volumes are specified. Total stops is a count of vehicle stops by SimTraffic. Whenever a vehicle s speed drops below 10 ft/s, a stop is added. A vehicle is considered moving again when it speed reaches 15 ft/s. Stops per vehicle is calculated by SimTraffic by dividing the number of stops by the number of vehicles in the system. Travel time is a total time each vehicle was present in the simulation area. Travel time in SimTraffic includes time spent by vehicles denied entry to the system. Due to the traffic balancing procedure used to generate the input volumes for the simulation, denied entry vehicles were not a factor in the travel time estimations and did not contribute to travel time in the analysis. Total vehicles entering network is the number of vehicles on the simulation network during the simulation, excluding entry intersection vehicles that do not travel on an approach to an intersection on the main parkway. Each vehicle is only counted once. A methodology was devised to provide a comparison of the Base Network intersections to the MLT intersections and the CFIs by individual traffic movement. This was particularly useful in evaluating the impacts of the alternative intersection designs on the left-turn and through movements. The process is illustrated through a series of exhibits. Exhibit 2-9 shows a standard intersection from the Base Network and identifies each approach movement by a number designation, and shows the same numbered movements in an associated MLT intersection. The numbers indicate how each movement tracks through the intersection. Reference letters are used to identify the individual movement locations within the MLT scheme. Exhibit 2-10 shows how data from a SimTraffic simulation of the standard intersection were compared to the MLT intersection by movement, by location within the intersection, and for the intersection scheme as a whole. Exhibit 2-11 provides the numbered movements by location for the CFI intersection and Exhibit 2-12 shows the accounting process for the delay estimates using a CFI. The CFIs were applied along the corridor at the following three locations: Parkway Parkway intersection at Mile 6. Parkway Major Arterial intersection at Mile 10. Parkway Minor Arterial intersection at Mile 7. The compilation of the performance metrics for the comparison of alternatives was conducted at following three levels of analysis: Aggregate results for all three intersections where application of a CFI was assumed. The individual CFI intersection locations by traffic movement. Phase 2 Continuous Flow Intersections 20

40 Exhibit 2-9 COMPARISON OF TRAFFIC MOVEMENTS THROUGH A STANDARD INTERSECTION AND AN MLT INTERSECTION Standard Intersection N LEGEND Turning Number MLT Intersection Configuration H G F 4+7 N K I J 2 5 D E C LEGEND A Turning Number Location Reference 1+10 L B A Phase 2 Continuous Flow Intersections 21

41 Exhibit 2-10 STUDY METHOD OF DELAY COMPARISON FOR AN MLT INTERSECTION Standard Intersection Delay Summary # Simulation Volume Delay / Vehicle (s/veh) Total Delay (veh-hr) EBL EBT EBR WBL WBT WBR NBL NBT 8 1, NBR SBL SBT 11 1, SBR TOTAL 7, Delay / Vehicle by by Location (seconds) Simulation Volume A B C D E F G H I J K L MLT Intersection Delay by Location (seconds) Additional Travel Time (s) Additional Travel Time + Delay / Total Delay (veh-hr) # Vehicle (s/veh) EBL EBT EBR WBL WBT WBR NBL NBT 8 1, NBR SBL SBT 11 1, SBR TOTAL 7, H G F N I J 6 53 K E 28 5 D C L B MLT Intersection Delay Summary 23 A Phase 2 Continuous Flow Intersections 22

42 Exhibit 2-11 TRAFFIC MOVEMENTS THROUGH A CONTINUOUS FLOW INTERSECTION CFI Intersection C H LEGEND 3 12 G O T R Y B M AA U X I K 8+1 W Z V BB 8 N 9 L Q S P E 6 9 F N D Turning Number J A A Location Reference Phase 2 Continuous Flow Intersections 23

43 Exhibit 2-12 STUDY METHOD OF DELAY COMPARISON FOR A CONTINOUS FLOW INTERSECTION CFI Intersection Delay by Location (seconds) # 7.4 B L K M 1.2 T H W 13.6 Q 0.1 AA 0.9 G U Z 1.8 P D C O 2.1 Y V E 0.9 BB 0.1 R 15.4 X F S N I J A 7.4 CFI Intersection Delay Summary Delay / Vehicle by by Location Simulation Volume A B C D E F G H I J K L M N O P Q R S T U V W X Y Z AA BB EBL EBT 2 2, EBR WBL WBT 5 1, WBR NBL NBT 8 1, NBR SBL SBT 11 2, SBR TOTAL 11, N Delay / Vehicle (s/veh) Total Delay (veh-hr) Phase 2 Continuous Flow Intersections 24

44

45 3. SUMMARY OF TRAFFIC VOLUME, SIGNAL TIMING, AND TRAFFIC SPEED INPUT VALUES The following provides a brief summary of the traffic volume, traffic signal timing, and traffic speed input values used in the analysis. Aside from the number of traffic lanes, traffic volume and signal timing are the two primary factors affecting the intersection levels of service and delay estimates. Traffic speed is important in establishing traffic signal coordination and traffic progression along the main parkway corridor. A summary of the traffic volume, traffic signal timing, and traffic speed input data are provided below. TRAFFIC VOLUME INPUT DATA There are 33 intersections in the simulation of the Base Network (excluding entry intersections), 52 in the Base Network with 3 CFI (excluding entry intersections), 204 intersections in the simulation of the MLT Network (excluding entry intersections), and 222 in the MLT Network with 3 CFIs (excluding entry intersections) so the presentation of the traffic volumes at each location, by traffic movement, is somewhat impractical. The information below provides the average, minimum, and maximum volumes by approach and movement for the main parkway, and for the main parkway cross streets by type of cross street. That is, the average, minimum, and maximum main parkway approach volumes are presented for each cross street type, and the average, minimum, and maximum approach volumes on each cross street type at the main parkway are also presented for each analysis scenario. BASE CAPACITY TRAFFIC VOLUMES The Base Capacity traffic volumes are presented in Exhibit 3-1. It should be noted that because SimTraffic uses a random simulation process in generating traffic volumes for the analysis, the actual volumes used in each simulation may be more or less than the values provided below. MLT CAPACITY VOLUMES The MLT Capacity volumes are provided in Exhibit 3-2. Exhibit 3-3 provides the traffic volumes used for the additional analysis of both the MLT parkway-parkway intersection and the SPUI parkway-parkway intersection scenarios for the MLT Uniform Capacity condition. CFI CAPACITY VOLUMES The CFI Capacity volumes are provided in Exhibit 3-4. Exhibit 3-5 provides the traffic volumes used for the additional analysis of both the CFI parkway-parkway intersection and the SPUI parkway-parkway intersection scenarios for the CFI Uniform Capacity condition. TRAFFIC SIGNAL TIMING INPUT VALUES A summary of typical traffic signal timing input data for the study intersections is provided in Exhibits 3-6 and 3-7. Typical timing splits (green + change + clearance intervals) are provided for the main parkway and for the cross streets by type of street for each analysis scenario conducted. Split values are provided for the left-turn movement (LT), the through and right-turn movement (Thru/RT), and the U-Turn movement (for the MLT alternatives only). The cycle length input for the main parkway intersections is also provided. Phase 2 Continuous Flow Intersections 25

46 TRAFFIC SPEEDS The traffic speed input for each roadway type is provided in Exhibit 3-8. This represents the speed limit on each roadway type and the average travel speed for drivers under free flow traffic conditions. Drivers in the simulation will travel faster or slower than the speed limit depending on the level of traffic congestion and the driver type speed factors used by SimTraffic. Phase 2 Continuous Flow Intersections 26

47 Left Turn Base Condition Traffic Volumes Input Parkway Volumes Average Minimum 1 Maximum 1 Right Turn Total U-Turn Left Turn Right Turn Total U-Turn Left Turn Right Turn Total U-Turn Cross Street Type Thru Thru Thru Parkway 187 3, , , , , , Major Arterial 210 3, , , , , , Minor Arterial 110 3, , , , , , Collector 62 4, , , , , , Left Turn Input Cross Street Volumes Average Minimum 1 Maximum 1 Right Turn Total U-Turn Left Turn Right Turn Total U-Turn Left Turn Right Turn Total U-Turn Cross Street Type Thru Thru Thru Parkway 211 1, , , , , , Major Arterial 184 1, ,834 N / A 146 1, ,650 N / A 247 1, ,045 N / A Minor Arterial ,113 N / A N / A ,544 N / A Collector N / A N / A N / A 1. volumes w ill not sum to Total volume. Exhibit 3-1 BASE CAPACITY INPUT TRAFFIC VOLUMES 1 Phase 2 Continuous Flow Intersections 27

48 Left Turn Phase 2 Continuous Flow Intersections 28 MLT Capacity Condition Traffic Volumes Input Parkway Volumes Average Minimum 1 Maximum 1 Right Turn Total U-Turn Left Turn Right Turn Total U-Turn Left Turn Right Turn Total U-Turn Cross Street Type Thru Thru Thru Parkway 187 3, , , , , , Major Arterial 210 3, , , , , , Minor Arterial 110 3, , , , , , Collector 67 4, , , , , , Left Turn Input Cross Street Volumes Average Minimum 1 Maximum 1 Right Turn Total U-Turn Left Turn Right Turn Total U-Turn Left Turn Right Turn Total U-Turn Cross Street Type Thru Thru Thru Parkway 211 1, , , , , , Major Arterial 184 1, ,827 N / A 146 1, ,650 N / A 247 1, ,045 N / A Minor Arterial ,079 N / A N / A ,200 N / A Collector N / A N / A N / A 1. volumes w ill not sum to Total volume. Exhibit 3-2 MLT CAPACITY INPUT TRAFFIC VOLUMES 1 Exhibit 3-3 PARKWAY-PARKWAY INTERSECTION ANALYSIS INPUT MLT UNIFORM CAPACITY VOLUMES 1 Parkway-Parkway Intersection Analysis Input Volumes (MLT Uniform Capacity) Left Turn Right Turn Total U-Turn Approach Thru Eastbound 461 2, , Westbound 415 1, , Northbound 390 1, , Southbound 367 2, , Volumes used for the MLT, SPUI, and CFI evaluations.

49 Exhibit 3-4 CFI CAPACITY INPUT TRAFFIC VOLUMES CFI Capacity Condition Traffic Volumes Input Parkway Volumes Left Turn Parkway 1 Cross Street 2 Right Left Right Thru Turn Total Turn Thru Turn Cross Street Type Total Parkway 187 3, , , ,890 Major Arterial 192 4, , , ,750 Minor Arterial 94 4, , , volumes are averaged from north and south approaches. 2. volumes are averaged from east and w est approaches. Exhibit 3-5 PARKWAY-PARKWAY INTERSECTION ANALYSIS INPUT CFI UNIFORM CAPACITY VOLUMES 1 Parkway-Parkway Intersection Analysis Input Volumes (CFI Uniform Capacity) Approach Left Turn Thru Right Turn Total U-Turn Eastbound 461 2, ,514 0 Westbound 415 2, ,577 0 Northbound 390 2, ,444 0 Southbound 367 2, , Volumes used for the CFI and SPUI evaluations. Phase 2 Continuous Flow Intersections 29

50 Exhibits 3-6 TYPICAL TRAFFIC SIGNAL TIMING BY ROADWAY TYPE Green + Change + Clearance Intervals (seconds) Parkway Cross Street Cycle Length Alternative Scenario Description LT Thru/RT U-turn Type LT Thru/RT U-Turn (seconds) Parkway Base Network with Base Capacity Volumes Major Arterial Minor Arterial Parkway MLT Network with Base Capacity Volumes Major Arterial 41 N / A N / A Minor Arterial 39 N / A / Parkway 58 / CFI Locations with Base Capacity Volumes 68 / Major Arterial 52 / / Minor Arterial Parkway MLT Network with MLT Capacity Volumes Major Arterial 28 N / A Minor Arterial 28 N / A / Parkway 32 / CFI Locations with MLT/CFI Capacity Volumes 52 / Major Arterial 28 / / Minor Arterial 27 / Main intersection / upstream left-turn time. Phase 2 Continuous Flow Intersections 30

51 Exhibit 3-7 TRAFFIC SIGNAL TIMING FOR THE PARKWAY-PARKWAY INTERSECTION ANALYSES MLT Parkway - Parkway Intersection Analysis (MLT Uniform Capacity Volumes) Green + Change + Clearance Intervals (seconds) N / S Parkway E / W Parkway Cycle Length Thru/RT U-Turn Thru/RT U-Turn (seconds) SPUI Parkway - Parkway Intersection Analysis (MLT Uniform Capacity Volumes) Green + Change + Clearance Intervals (seconds) Cycle Length E / W Thru/RT E / W LT N / S LT (seconds) CFI Parkway - Parkway Intersection Analysis (MLT/CFI Uniform Capacity Volumes) Green + Change + Clearance Intervals (seconds) N / S Parkway E / W Parkway Cycle Length Thru/LT Upstream LT Thru/LT Upstream LT (seconds) SPUI Parkway - Parkway Intersection Analysis (CFI Uniform Capacity Volumes) Green + Change + Clearance Intervals (seconds) Cycle Length E / W Thru/RT E / W LT N / S LT (seconds) SPUI Parkway - Parkway Intersection Analysis (MLT Network Base Capacity Volumes) Green + Change + Clearance Intervals (seconds) Cycle Length E / W Thru/RT E / W LT N / S LT (seconds) SPUI Parkway - Parkway Intersection Analysis (MLT Network MLT Capacity Volumes) Green + Change + Clearance Intervals (seconds) Cycle Length E / W Thru/RT E / W LT N / S LT (seconds) Phase 2 Continuous Flow Intersections 31

52 Exhibit 3-8 INPUT ROADWAY TRAVEL SPEED BY ROADWAY TYPE Roadway Type Link Speed (mph) Parkway 45 Major Arterial 40 Minor Arterial 35 Collector 30 Phase 2 Continuous Flow Intersections 32

53 4. ANALYSIS RESULTS The simulation results are presented through a series of tables and charts for each analysis scenario. The simulation results represent average values of the performance metrics from five random simulations of each scenario that was tested. Each simulation represents a one-hour time period of traffic flow. The aggregate results for the study intersections are presented through a summary of the performance metrics that are used to compare corridor design alternatives under a given set of traffic volume assumptions. These summary statistics provide the following: Total corridor delay in hours. Corridor delay per vehicle in seconds. Total corridor stops. Corridor stops per vehicle. Total corridor travel time in hours. Total vehicles entering the intersection or intersections being evaluated. The traffic operations at individual intersections by traffic movement are also compared under each analysis scenario. BASE CAPACITY TRAFFIC VOLUME SCENARIO The Base Capacity scenario represents a traffic condition where the Base Network design is operating at or near capacity. The maximum hourly approach volumes used in the Base Capacity analysis (3,100 vehicles per hour) represents an average daily traffic of approximately 62,000 vehicles per day for the main parkway assuming the peak-hour is ten percent of the daily traffic. This value is consistent with values provided in the State of Florida Quality/Level of Service Handbook for an 8-lane divided roadway in urban areas 6. Considering only the through movement at signalized intersections, the main parkway capacity is approximately 600 to 675 vehicles per hour per through lane for the Base Network, under the assumptions made in the previous study. The Base Capacity traffic volumes were used to evaluate and compare the traffic operations of the following design alternatives: Base Network design. Base Network design with three continuous flow intersections: - Parkway-Parkway intersection (Mile 6). - Parkway-Major Arterial intersection (Mile 10). - Parkway-Minor Arterial intersection (Mile 7) with partial CFI (parkway approaches only). MLT Network design. MLT Network design with a SPUI at the Parkway-Parkway intersection (Mile 6). 6 State of Florida, Department of Transportation, Quality/Level of Service Handbook, 2002, page 85. Note that the Florida capacity values assume that the peak-hour is 9.5 percent of the daily traffic. Phase 2 Continuous Flow Intersections 33

54 The comparison of results is provided at the following two levels of analysis: Aggregate results for three intersections (Mile 6, Mile 7, and Mile 10). Individual intersection results. Aggregate Results for Three Study Intersections with Base Capacity Volumes The aggregate results for the three study intersections are presented in Exhibit 4-1. It should be noted that only a partial CFI was used with the Base Capacity volumes at the minor arterial intersection (Mile 7) because this is all that was required to show significant improvement in traffic operations with the Base Capacity volumes. The results indicate: The MLT intersections reduce delay by 33 percent, stops by 19 percent, and total travel time through the three intersections by 15 percent in comparison to the Base intersections. The CFI intersections reduce delay by 46 percent, stops by 50 percent, and total travel time through the three intersections by 46 percent in comparison to the Base intersections. The CFI intersections show a substantial improvement in traffic operations over the MLT intersections by reducing delay by an additional 19 percent, stops by 38 percent, and travel time by 36 percent. Exhibit 4-1 AGGREGATE INTERSECTION PERFORMANCE MEASURES WITH BASE CAPACITY VOLUMES Base Capacity Volumes Mile 6, 7, & 10 Intersections Comparison Performance Measure Base MLT 2 Full/ 1 Partial CFI Base to MLT % Diff. Base to CFI % Diff. MLT to CFI % Diff. Total Delay (hours) % -46.1% -19.0% Delay per Vehicle (seconds) % -46.3% -19.2% Total Stops 28,007 22,703 14, % -49.6% -37.8% Stops per Vehicle % -49.7% -37.9% Total Travel Time (hours) Total Vehicles Entering Intersections % -45.7% -35.9% 28,686 28,708 28, % 0.3% 0.2% Parkway-Parkway Intersection (Mile 6) with Base Capacity Volumes The intersection at Mile 6 of the corridor is the intersection of the main parkway with the eastwest parkway. The MLT intersection assumed the provision of a U-turn opportunity on each intersection approach consistent with the overall parkway design for the MLT condition. A SPUI intersection design was also simulated at this intersection in the work conducted for the previous study. A CFI was applied in this study. A summary table of the performance measures is provided in Exhibit 4-2. The results indicate: Phase 2 Continuous Flow Intersections 34

55 The MLT design reduces intersection delay by 39 percent, vehicle stops by 30 percent, and travel time in the intersection by 17 percent in comparison to the Base design with the Base Capacity volumes. The CFI design reduces intersection delay by 59 percent, vehicle stops by 69 percent, and travel time by 58 percent in comparison to the Base design with the Base Capacity volumes. The CFI provides a substantial improvement over the MLT design by reducing delay by 32 percent, stops by 55 percent, and travel time by 50 percent in comparison to the MLT configuration. The SPUI reduces delay 59 percent, stops by 46 percent, and travel time by 26 percent in comparison to the Base design. The SPUI design provides a level of delay reduction over the Base design that is virtually the same as the CFI, but the number of stops and total travel time are higher than the CFI design. This latter condition is due primarily to the oversaturated high-volume northbound and southbound right-turn movements (see Exhibit 4-3), for which only a single right-turn lane was provided on each approach. Exhibit 4-2 INTERSECTION PERFORMANCE MEASURES FOR THE PARKWAY-PARKWAY (MILE 6) INTERSECTION WITH BASE CAPACITY VOLUMES Base Capacity Volumes Mile 6 Intersection Comparison Performance Measure Base MLT CFI SPUI Base to MLT % Diff. Base to CFI % Diff. Base to SPUI % Diff. MLT to CFI % Diff. MLT to SPUI % Diff. CFI to SPUI % Diff Total Delay (hours) % -58.7% -59.3% -32.6% -33.6% -1.4% Delay per Vehicle (seconds) % -58.8% -59.3% -32.2% -33.1% -1.3% Total Stops 12,582 8,864 3,934 4, % -68.7% -61.7% -55.6% -45.7% 22.4% Stops per Vehicle % -68.8% -61.8% -55.4% -45.3% 22.6% Total Travel Time (hours) Total Vehicles Entering Intersection % -58.5% -38.5% -50.0% -25.9% 48.3% 11,372 11,471 11,403 11, % 0.3% 0.1% -0.6% -0.8% -0.2% Exhibit 4-3 provides a comparison of the delay by movement for each of the intersection design alternatives. Examination of the information contained in Exhibit 4-3 indicates the following: The strength of the MLT design is in the reduction in delay to the through and right-turn movements at the intersection resulting from the two-phase signalization. The MLT design increases delay to the left-turn movement due to the fact that each leftturn vehicle must travel through the intersection twice. A significant portion of the delay is the increase in travel time associated with the U-turn movements, which was estimated at an additional 20 seconds for this study based on the distance to the U-turn opportunity and the link travel speed. Phase 2 Continuous Flow Intersections 35

56 Exhibit 4-3 PARKWAY-PARKWAY INTERSECTION (Mile 6) DELAY PER MOVEMENT BY INTERSECTION DESIGN ALTERNATIVE(Base Capacity Volumes) Parkway / Parkway (Mile 6) Intersection Delay per Vehicle by Base Configuration with Base Capacity Volumes Parkway / Parkway (Mile 6) Intersection Delay per Vehicle by MLT Configuration with Base Capacity Volumes Delay / Veh (sec) Total Delay = 214 Hours Delay / Vehicle = 68 Seconds Total Stops = 12,582 Stops / Vehicle = 1.11 Total Travel Time = 329 Hours Total Vehicles = 11, , , , , Hourly 206 Flow Rate (vph) Delay / Veh (sec) Total Delay = 131 Hours Delay / Vehicle = 41 Seconds Total Stops = 8,864 Stops / Vehicle = 0.77 Total Travel Time = 273 Hours Total Vehicles = 11, , , , , RT Lane UT Lane Other Delay Hourly 207 Flow Rate (vph) EBL EBT EBR WBL WBT WBR NBL NBT NBR SBL SBT SBR 0 EBL EBT EBR WBL WBT WBR NBL NBT NBR SBL SBT SBR Parkway / Parkway (Mile 6) Intersection Delay per Vehicle by CFI Configuration with Base Capacity Volumes Parkway / Parkway (Mile 6) Intersection Delay per Vehicle by SPUI Configuration with Base Capacity Volumes Total Delay = 88 hours Delay per Vehicle = 28 seconds Total Stops = 3,934 Stops per Vehicle = 0.34 Travel Time = 136 hours Total Vehicles = 11, Total Delay = 87 Hours Delay / Vehicle = 28 Seconds Total Stops = 4,815 Stops / Vehicle = 0.42 Total Travel Time = 202 Hours Total Vehicles = 11, Delay / Veh (sec) , , , , RT Lane LT Lane Other Delay Hourly 204 Flow Rate (vph) Delay / Veh (sec) , , , , Hourly 199 Flow Rate (vph) EBL EBT EBR WBL WBT WBR NBL NBT NBR SBL SBT SBR 0 EBL EBT EBR WBL WBT WBR NBL NBT NBR SBL SBT SBR Phase 2 Continuous Flow Intersections 36

57 The strength of the CFI is that it reduces delay to all movements in the intersection. The two-phase signal timing reduces delay to both the through movement and the left-turn movements, while the free right-turn significantly reduces delay to the right-turn movements. The free right-turn configuration is capable of accommodating high volumes in a single right-turn lane. While the CFI delay to the through movement is similar to that of the MLT design, the delays to the left-turn and right-turn movements are significantly lower. Therefore, the CFI will accommodate higher volume turn movements than the MLT design at an acceptable level of delay. The major strength of the SPUI design results from the grade separation of the highest volume through movements. The three-phase traffic signal also reduces delay to those movements passing through the signal in comparison to the Base Design. The SPUI provided the lowest levels of delay to the left-turn movements for the volumes used in the study. Delays to the eastbound and westbound through movements passing through the traffic signal were comparable to the delays in the MLT design, but higher than those in the CFI design. The weakness of the SPUI in this test case was the delay to the high-volume northbound and southbound right-turn movements. In practice, this weakness would be overcome through the use of dual right-turn lanes for these movements. Parkway-Major Arterial Intersection (Mile 10) with Base Capacity Volumes The intersection at Mile 10 of the corridor is the intersection of the main parkway with the eastwest major arterial roadway. The MLT intersection assumed the provision of a U-turn opportunity on the north and south parkway legs of the intersection consistent with the overall parkway design for the MLT condition. A full CFI, with CFI left and right-turn treatments on each approach, was applied in this study. A summary table of the performance measures for each intersection design alternative is provided in Exhibit 4-4. The results indicate: The MLT design reduces intersection delay by 23 percent, and travel time in the intersection by 11 percent in comparison to the Base design with the Base Capacity volumes. Stops remained the same in the MLT intersection compared to the Base design. The CFI design reduces intersection delay by 42 percent, vehicle stops by 41 percent, and travel time by 49 percent in comparison to the Base design with the Base Capacity volumes. The CFI provides a substantial improvement over the MLT design by reducing delay by 25 percent, stops by 42 percent, and travel time by 43 percent in comparison to the MLT configuration. The CFI improvement in traffic operations in comparison to the MLT design is slightly less at the major arterial intersection than at the parkway-parkway intersection. Phase 2 Continuous Flow Intersections 37

58 Exhibit 4-4 INTERSECTION PERFORMANCE MEASURES FOR THE PARKWAY-MAJOR ARTERIAL (MILE 10) INTERSECTION WITH BASE CAPACITY VOLUMES Base Capacity Volumes Mile 10 Intersection Comparison Performance Measure Base MLT CFI Base to MLT % Diff. Base to CFI % Diff. MLT to CFI % Diff. Total Delay (hours) % -42.2% -25.0% Delay per Vehicle (seconds) % -42.4% -25.6% Total Stops 7,780 7,925 4, % -41.0% -42.1% Stops per Vehicle % -41.2% -42.6% Total Travel Time (hours) Total Vehicles Entering Intersection % -49.2% -42.9% 9,119 9,074 9, % 0.4% 0.8% Exhibit 4-5 provides a comparison of the delay by movement for each of the intersection design alternatives at the major arterial intersection. Examination of the information contained in Exhibit 4-5 indicates the following: Both the MLT intersection and the CFI provide significant improvement over the Base design through the reduction in delay for the through movement. The CFI is slightly better in this regard because the left-turn traffic does not pass through the main intersection with the through traffic. The main weakness of the MLT design is the delay associated with the left-turn movement. The MLT design increases delay to the left-turn movement due to the fact that each left-turn vehicle must travel through the intersection twice. A significant portion of the delay is the increase in travel time associated with the U-turn movements, which was estimated at an additional 20 seconds for this study based on the distance to the U- turn opportunity and the link travel speed. The major strength of the CFI is that it reduces delay for all movements in comparison to both the Base and MLT design. While the CFI delay to the through movement is slightly lower than that with MLT design, the delays to the left-turn and right-turn movements are significantly lower. Therefore, the CFI will accommodate higher volume turn movements than the MLT design at an acceptable level of delay. The free right-turn feature provides a significant delay reduction. Phase 2 Continuous Flow Intersections 38

59 Exhibit 4-5 PARKWAY-MAJOR ARTERIAL INTERSECTION (Mile 10) DELAY PER MOVEMENT BY INTERSECTION DESIGN ALTERNATIVE (Base Capacity Volumes) Parkway / Major Arterial (Mile 10) Intersection Delay per Vehicle by Base Configuration with Base Capacity Volumes Parkway / Major Arterial (Mile 10) Intersection Delay per Vehicle by MLT Configuration with Base Capacity Volumes Total Delay = 133 Hours Delay / Vehicle = 53 Seconds Total Stops = 7,780 Stops / Vehicle = 0.85 Total Travel Time = 244 Hours Total Vehicles = 9, Total Delay = 103 Hours Delay / Vehicle = 41 Seconds Total Stops = 7,925 Stops / Vehicle = 0.87 Total Travel Time = 217 Hours Total Vehicles = 9, Delay / Veh (sec) , , , , Hourly Flow Rate (vph) Delay / Veh (sec) , , , , RT Lane UT Lane Other Delay Hourly 170 Flow Rate (vph) EBL EBT EBR WBL WBT WBR NBL NBT NBR SBL SBT SBR 0 EBL EBT EBR WBL WBT WBR NBL NBT NBR SBL SBT SBR Parkway / Major Arterial (Mile 10) Intersection Delay per Vehicle by CFI Configuration with Base Capacity Volumes Delay / Veh (sec) Total Delay = 77 hours Delay per Vehicle = 30 seconds Total Stops = 4,590 Stops per Vehicle = 0.50 Travel Time = 124 hours Total Vehicles = 9,151 2,314 2, , , EBL EBT EBR WBL WBT WBR NBL NBT NBR SBL SBT SBR RT Lane LT Lane Other Delay Hourly 204 Flow Rate (vph) Phase 2 Continuous Flow Intersections 39

60 Parkway-Minor Arterial Intersection (Mile 7) with Base Capacity Volumes The intersection at Mile 7 of the corridor is the intersection of the main parkway with the eastwest minor arterial roadway. The MLT intersection assumed the provision of a U-turn opportunity on the north and south parkway legs of the intersection consistent with the overall parkway design for the MLT condition. Only a partial CFI, with the CFI left and right-turn treatments on the north-south parkway approaches, was applied with the Base Capacity volumes at this location. This was done to test the traffic operations characteristics of this alternative CFI design treatment. With a partial CFI the minor cross street approaches are the same as the standard Base design, with separate phases for the through and left-turn movements. Therefore, the partial CFI requires a threephase signal treatment. A summary table of the performance measures for each intersection design alternative is provided in Exhibit 4-6. The results indicate: The MLT design reduces intersection delay by 35 percent, stops by 23 percent, and travel time in the intersection by 17 percent in comparison to the Base design with the Base Capacity volumes. The partial CFI design reduces intersection delay by 29 percent, vehicle stops by 27 percent, and travel time by 26 percent in comparison to the Base design with the Base Capacity volumes. The partial CFI provides improvements in traffic operations that are generally comparable to the MLT design at this location under the assumed traffic volumes. Exhibit 4-6 INTERSECTION PERFORMANCE MEASURES FOR THE PARKWAY-MINOR ARTERIAL (MILE 7) INTERSECTION WITH BASE CAPACITY VOLUMES Base Capacity Volumes Mile 7 Intersection Comparison Performance Measure Base MLT Partial CFI Base to MLT % Diff. Base to CFI % Diff. MLT to CFI % Diff. Total Delay (hours) % -28.7% 10.4% Delay per Vehicle (seconds) % -28.9% 9.8% Total Stops 7,645 5,914 5, % -26.8% -5.4% Stops per Vehicle % -26.9% -5.9% Total Travel Time (hours) Total Vehicles Entering Intersection % -26.2% -11.0% 8,195 8,163 8, % 0.2% 0.6% Exhibit 4-7 provides a comparison of the delay by movement for each of the intersection design alternatives at the major arterial intersection. Examination of the information contained in Exhibit 4-7 indicates the following: Phase 2 Continuous Flow Intersections 40

61 The MLT design provides significant reduction in delay to both the through and right-turn movements in comparison to the Base design, but the delay to the left-turn movements is higher than for the Base design. The left-turn volumes are lower at this intersection than at either the parkway-parkway or parkway-major arterial intersections. The delay for the north-south through movements at the partial CFI are comparable to that of the MLT design and the delay to the north-south left-turns are lower than for either the Base design or the MLT design. The delays for the east-west through and left-turn movements in the partial CFI are comparable to that of the Base design. The delay for the east-west through movement in the partial CFI is higher than for the MLT design, due primarily to the additional signal phase with the partial CFI. Delay to the right-turn movements in the partial CFI are lower than for the Base design and the MLT design. Overall the partial CFI would provide a reasonable alternative to the Base design under the assumed traffic volumes and turn movements. Phase 2 Continuous Flow Intersections 41

62 Exhibit 4-7 PARKWAY-MINOR ARTERIAL INTERSECTION (Mile 7) DELAY PER MOVEMENT BY INTERSECTION DESIGN ALTERNATIVE (Base Capacity Volumes) Delay / Veh (sec) Parkway / Minor Arterial (Mile 7) Intersection Delay per Vehicle by Base Configuration with Base Capacity Volumes Total Delay = 125 Hours Delay / Vehicle = 55 Seconds Total Stops = 7,645 Stops / Vehicle = 0.93 Total Travel Time = 261 Hours Total Vehicles = 8, o u r l Hourly Flow Rate (vph) Delay / Veh (sec) Parkway / Minor Arterial (Mile 7) Intersection Delay per Vehicle by MLT Configuration with Base Capacity Volumes Total Delay = 81 Hours Delay / Vehicle = 36 Seconds Total Stops = 5,914 Stops / Vehicle = 0.72 Total Travel Time = 216 Hours Total Vehicles = 8,163 RT Lane UT Lane Other Delay ,376 EBL EBT EBR WBL WBT WBR NBL NBT NBR SBL SBT SBR , , EBL EBT EBR WBL WBT WBR NBL NBT NBR SBL SBT SBR 89 2, Hourly 73 Flow Rate (vph) Parkway / Minor Arterial (Mile 7) Intersection Delay per Vehicle by Partial CFI Configuration with Base Capacity Volumes Delay / Veh (sec) Total Delay = 89 hours Delay per Vehicle = 39 seconds Total Stops = 5,597 Stops per Vehicle = 0.68 Travel Time = 192 hours Total Vehicles = 8, , , EBL EBT EBR WBL WBT WBR NBL NBT NBR SBL SBT SBR RT Lane LT Lane Other Delay Hourly 204 Flow Rate (vph) Phase 2 Continuous Flow Intersections 42

63 MLT CAPACITY TRAFFIC VOLUME SCENARIO The MLT Capacity scenario represents a traffic condition where the MLT network is generally operating at capacity. The maximum hourly approach volumes used in the MLT Capacity analysis (4,500 vehicles per hour) represents an average daily traffic of approximately 90,000 vehicles per day for the main parkway assuming the peak-hour is ten percent of the daily traffic. This result may be somewhat conservative in that the capacity of the right-turn lanes was not explored in this or the previous study. Considering only the through movement at signalized intersections, the MLT design capacity is approximately 975 to 1,025 vehicles per hour per through lane. Typically, right-turn lane capacity is approximately 85 to 95 percent of through lane capacity. U-turn lane capacity under stop control or traffic signal control was not evaluated as an element of this or the previous study. The MLT Capacity traffic volumes were used to evaluate and compare the traffic operations of the following design alternatives: MLT Network design with three continuous flow intersections: - Parkway-Parkway intersection (Mile 6). - Parkway-Major Arterial intersection (Mile 10). - Parkway-Minor Arterial intersection (Mile 7) with partial CFI (parkway approaches only). MLT Network design with a SPUI at the Parkway-Parkway intersection (Mile 6). The comparison of results is provided at the following two levels of analysis: Aggregate results for three intersections (Mile 6, Mile 7, and Mile 10) for the MLT design and for the CFI design. Individual intersection results. Aggregate Results for Three Study Intersections with MLT Capacity Volumes The aggregate results for the three study intersections are presented in Exhibit 4-8. It should be noted that a full CFI was used with the MLT Capacity volumes at the minor arterial intersection (Mile 7), so that a full CFI was applied to all three intersections. The results indicate: The CFIs reduce delay by 22 percent, and stops by 28 percent, and travel time by 18 percent in comparison to the MLT design. The reduction in delay resulting from the CFIs is slightly greater with the MLT Capacity volumes than with the Base Capacity volumes, but the reduction in stops and travel time is less than with the Base Capacity volumes. Phase 2 Continuous Flow Intersections 43

64 Exhibit 4-8 AGGREGATE INTERSECTION PERFORMANCE MEASURES WITH MLT CAPACITY VOLUMES MLT Capacity Volumes Mile 6, 7, & 10 Intersections Comparison Performance Measure MLT 3 Full CFI MLT to CFI % Diff. Total Delay (hours) % Delay per Vehicle (seconds) % Total Stops 48,226 34, % Stops per Vehicle % Total Travel Time (hours) Total Vehicles Entering Intersections 1, % 35,903 35, % Parkway-Parkway Intersection (Mile 6) with MLT Capacity Volumes The intersection at Mile 6 of the corridor is the intersection of the main parkway with the eastwest parkway. The MLT intersection assumed the provision of a U-turn opportunity on each intersection approach consistent with the overall parkway design for the MLT condition. A SPUI intersection design was also simulated at this intersection in the work conducted for the previous study. A CFI was applied in this study. A summary table of the performance measures is provided in Exhibit 4-9. The results indicate: The CFI reduced delay by 14 percent, stops by 20 percent, and travel time by 5 percent in comparison to the MLT intersection. This is a substantially lower reduction in these parameters than simulated with the Base Capacity Volumes, suggesting that the CFI loses some advantage over the MLT intersection as the volume of through vehicles increases. The SPUI reduced delay by 64 percent, stops by 76 percent, and travel time by 42 percent in comparison to the MLT intersection. This is substantially better than the results using the Base Capacity volumes because more vehicles are traveling through the grade separation portion of the intersection with the MLT Capacity volumes. The SPUI reduced delay by 58 percent, stops by 70 percent, and travel time by 39 percent in comparison to the CFI intersection. This is the opposite result obtained from the simulations with the Base Capacity volumes, where the CFI produced lower stops and travel time than the SPUI. This suggests that the advantage of the CFI is lost in comparison to the SPUI when the through traffic on the grade separation in the SPUI reaches a certain level somewhere between the Base Capacity volume and the MLT Capacity volume. Phase 2 Continuous Flow Intersections 44

65 Exhibit 4-9 INTERSECTION PERFORMANCE MEASURES FOR THE PARKWAY-PARKWAY (MILE 6) INTERSECTION WITH MLT CAPACITY VOLUMES MLT Capacity Volumes Mile 6 Intersection Comparison Performance Measure MLT Full CFI SPUI MLT to CFI % Diff. MLT to SPUI % Diff. CFI to SPUI % Diff Total Delay (hours) % -64.5% -58.5% Delay per Vehicle (seconds) % -64.0% -58.0% Total Stops 21,282 16,833 5, % -76.0% -69.7% Stops per Vehicle % -75.6% -69.3% Total Travel Time (hours) Total Vehicles Entering Intersection % -42.4% -39.2% 13,778 13,709 13, % -1.6% -1.1% Exhibit 4-10 provides a comparison of the delay by movement for each of the intersection design alternatives at the parkway-parkway intersection. Examination of the information contained in Exhibit 4-10 indicates the following: While the through movements in the MLT design are at capacity, it is the left-turn movements that suffer the highest levels of delay. The greatest proportion of delay to the left-turn movements is generated as this traffic passes through the main intersection twice. The CFI significantly improves the delay to the left-turn movements over the MLT design, but the through movements on three approaches operate at capacity. The free right-turn movements in the CFI provide a substantial delay reduction over the MLT design. The SPUI provides significant delay reduction to all movements in the intersection. In this particular case the left-turn movements suffer less delay than with the CFI design, and the delay to the right-turn movements is similar to that of the CFI. The grade separated northbound and southbound through movements have virtually no delay. Phase 2 Continuous Flow Intersections 45

66 Exhibit 4-10 PARKWAY-PARKWAY INTERSECTION (Mile 6) DELAY PER MOVEMENT BY INTERSECTION DESIGN ALTERNATIVE (MLT Capacity Volumes) Parkway / Parkway (Mile 6) Intersection Delay per Vehicle by MLT Configuration with MLT Capacity Volumes Parkway / Parkway (Mile 6) Intersection Delay per Vehicle by CFI Configuration with MLT Capacity Volumes Total Delay = 284 Hours Delay / Vehicle = 74 Seconds Total Stops = 21,282 Stops / Vehicle = 1.54 Total Travel Time = 440 Hours Total Vehicles = 13, Total Delay = 242 Hours Delay / Vehicle = 64 Seconds Total Stops = 16,833 Stops / Vehicle = 1.23 Total Travel Time = 417 Hours Total Vehicles = 13,709 Delay / Veh (sec) , , , , EBL EBT EBR WBL WBT WBR NBL NBT NBR SBL SBT SBR RT Lane UT Lane Other Delay Hourly 206 Flow Rate (vph) Delay / Veh (sec) , , , ,246 EBL EBT EBR WBL WBT WBR NBL NBT NBR SBL SBT SBR RT Lane LT Lane Other Delay Hourly 205 Flow Rate (vph) Parkway / Parkway (Mile 6) Intersection Delay per Vehicle by SPUI Configuration with MLT Capacity Volumes 150 Total Delay = 101 Hours Delay / Vehicle = 27 Seconds Total Stops = 5,104 Stops / Vehicle = 0.38 Total Travel Time = 253 Hours Total Vehicles = 13,559 Delay / Veh (sec) Hourly Flow Rate (vph) , , , , EBL EBT EBR WBL WBT WBR NBL NBT NBR SBL SBT SBR Phase 2 Continuous Flow Intersections 46

67 Parkway-Major Arterial Intersection (Mile 10) with MLT Capacity Volumes The intersection at Mile 10 of the corridor is the intersection of the main parkway with the eastwest major arterial roadway. The MLT intersection assumed the provision of a U-turn opportunity on the north and south parkway legs of the intersection consistent with the overall parkway design for the MLT condition. A full CFI, with CFI left and right-turn treatments on each approach, was applied in this study. A summary table of the performance measures for each intersection design alternative is provided in Exhibit The results indicate: The CFI represents a significant improvement in traffic operations over the MLT design under the assumed traffic volume conditions. The CFI reduces delay by 36 percent, stops by 41 percent, and travel time by 29 percent in comparison to the MLT intersection. Exhibit 4-11 INTERSECTION PERFORMANCE MEASURES FOR THE PARKWAY-MAJOR ARTERIAL (MILE 10) INTERSECTION WITH MLT CAPACITY VOLUMES MLT Capacity Volumes Mile 10 Intersection Comparison Performance Measure MLT Full CFI MLT to CFI % Diff. Total Delay (hours) % Delay per Vehicle (seconds) % Total Stops 16,301 9, % Stops per Vehicle % Total Travel Time (hours) Total Vehicles Entering Intersection % 11,585 11, % Exhibit 4-12 provides a comparison of the delay by movement for each of the intersection design alternatives at the major arterial intersection. Examination of the information contained in Exhibit 4-12 indicates the following: The delay to the eastbound, westbound, and northbound through movements with the MLT design and the CFI design are approximately the same. The delay to the southbound through movement with the MLT design is approximately twice that of the southbound through movement in the CFI design. The delay to the left-turn movements in the MLT design are substantially higher than that simulated for the CFI design. The delay to the right-turn movements in the CFI design are also substantially lower than that simulated for the MLT design. Phase 2 Continuous Flow Intersections 47

68 Exhibit 4-12 PARKWAY-MAJOR ARTERIAL INTERSECTION (Mile 10) DELAY PER MOVEMENT BY INTERSECTION DESIGN ALTERNATIVE (MLT Capacity Volumes) Parkway / Major Arterial (Mile 10) Intersection Delay per Vehicle by MLT Configuration with MLT Capacity Volumes Delay / Veh (sec) Total Delay = 209 Hours Delay / Vehicle = 65 Seconds Total Stops = 16,301 Stops / Vehicle = 1.41 Total Travel Time = 350 Hours Total Vehicles = 11, , , ,450 EBL EBT EBR WBL WBT WBR NBL NBT NBR SBL SBT SBR , RT Lane UT Lane Other Delay Hourly 168 Flow Rate (vph) Parkway / Major Arterial (Mile 10) Intersection Delay per Vehicle by CFI Configuration with MLT Capacity Volumes Delay / Veh (sec) Total Delay = 134 Hours Delay / Vehicle = 41 Seconds Total Stops = 9,580 Stops / Vehicle = 0.83 Total Travel Time = 249 Hours Total Vehicles = 11, , , , , EBL EBT EBR WBL WBT WBR NBL NBT NBR SBL SBT SBR RT Lane LT Lane Other Delay Hourly 170 Flow Rate (vph) Phase 2 Continuous Flow Intersections 48

69 Parkway-Minor Arterial Intersection (Mile 7) with MLT Capacity Volumes The intersection at Mile 7 of the corridor is the intersection of the main parkway with the eastwest minor arterial roadway. The MLT intersection assumed the provision of a U-turn opportunity on the north and south parkway legs of the intersection consistent with the overall parkway design for the MLT condition. A full CFI, with the CFI left and right-turn treatments on all four approaches, was applied with the MLT Capacity volumes at this location. This was done to provide the best possible traffic operations characteristics of this CFI design treatment. The full CFI allows two-phase traffic signal operation. A summary table of the performance measures for each intersection design alternative is provided in Exhibit The results indicate: The CFI design reduces delay by 15 percent, stops by 20 percent, and travel time by 25 percent in comparison to the MLT intersection design. The full CFI used in this analysis with the MLT Capacity volumes produces much better results relative to the MLT design than the partial CFI design used with the Base Capacity volumes. Exhibit 4-13 INTERSECTION PERFORMANCE MEASURES FOR THE PARKWAY-MINOR ARTERIAL (MILE 7) INTERSECTION WITH MLT CAPACITY VOLUMES MLT Capacity Volumes Mile 7 Intersection Comparison Performance Measure MLT Full CFI MLT to CFI % Diff. Total Delay (hours) % Delay per Vehicle (seconds) % Total Stops 10,643 8, % Stops per Vehicle % Total Travel Time (hours) Total Vehicles Entering Intersection % 10,540 10, % Exhibit 4-14 provides a comparison of the delay by movement for each of the intersection design alternatives at the minor arterial intersection. Examination of the information contained in Exhibit 4-14 indicates the following: The delay to the through movements is essentially the same for both intersection designs. The CFI significantly reduces the delay to the left-turn movements in comparison to the MLT design. The CFI also reduces the delay to the right-turn movements in comparison to the MLT design. Phase 2 Continuous Flow Intersections 49

70 Exhibit 4-14 PARKWAY-MINOR ARTERIAL INTERSECTION (Mile 7) DELAY PER MOVEMENT BY INTERSECTION DESIGN ALTERNATIVE (MLT Capacity Volumes) Parkway / Minor Arterial (Mile 7) Intersection Delay per Vehicle by MLT Configuration with MLT Capacity Volumes Delay / Veh (sec) Total Delay = 140 Hours Delay / Vehicle = 48 Seconds Total Stops = 10,643 Stops / Vehicle = 1.01 Total Travel Time = 302 Hours Total Vehicles = 10, , , EBL EBT EBR WBL WBT WBR NBL NBT NBR SBL SBT SBR 72 RT Lane UT Lane Other Delay Hourly Flow Rate (vph) Parkway / Minor Arterial (Mile 7) Intersection Delay per Vehicle by CFI Configuration with MLT Capacity Volumes Delay / Veh (sec) Total Delay = 119 Hours Delay / Vehicle = 41 Seconds Total Stops = 8,503 Stops / Vehicle = 0.81 Total Travel Time = 228 Hours Total Vehicles = 10, , ,889 EBL EBT EBR WBL WBT WBR NBL NBT NBR SBL SBT SBR RT Lane LT Lane Other Delay Hourly Flow Rate (vph) Phase 2 Continuous Flow Intersections 50

71 PARKWAY-PARKWAY INTERSECTION (Mile 6) ANALYSIS ASSUMING UNIFORM APPROACH VOLUMES An analysis was conducted to evaluate the traffic operations at the parkway-parkway intersection assuming relatively balanced, or uniform, approach volumes that approximated the capacity of the MLT intersection design. The assumed volumes by approach and movement were provided previously in Exhibit 3-3. These volumes were developed in order to test the operating conditions of the intersections under conditions that require a nearly equally sharing of available green time by conflicting approaches. In addition, the assumed volumes represent relatively high volumes of right and left-turn movements. It should be noted that the MLT Uniform Capacity volumes represent traffic demand entering the intersection that is approximately 15 percent less than the traffic volume for the MLT Capacity volumes shown in Exhibit This indicates that the uniform approach volumes significantly reduce the overall capacity of the intersection. The MLT, SPUI, and CFI intersection designs were tested and compared using the assumed volumes. A summary table of the performance metrics for the three intersections types is provided in Exhibit These results indicate: The CFI significantly improves intersection performance over the MLT intersection design. The CFI reduces delay by 55 percent, stops by 60 percent, and travel time by 31 percent in comparison to the MLT design. The CFI and the SPUI provide improvements in traffic delay and travel time that are similar in magnitude in comparison to the MLT design. The SPUI reduces delay by 57 percent, stops by 73 percent, and travel time by 39 percent in comparison to the MLT design. The SPUI reduces delay by only 5 percent and travel time by only 11 percent in comparison to the CFI intersection. The SPUI reduces stops by 32 percent in comparison to the CFI primarily because of the large volume of traffic that passes through the intersection on the grade separation and does not stop. Exhibit 4-15 INTERSECTION PERFORMANCE MEASURES FOR THE PARKWAY-PARKWAY (MILE 6) INTERSECTION WITH MLT UNIFORM CAPACITY VOLUMES Parkway-Parkway Intersection MLT Uniform Capacity Volumes Intersection Type Comparison MLT Single U-Turn Lanes CFI SPUI MLT to CFI % Diff. MLT to SPUI % Diff. CFI to SPUI % Diff Performance Measure Total Delay (hours) % -57.8% -7.3% Delay per Vehicle (seconds) % -57.2% -5.3% Total Stops 13,754 5,530 3, % -73.0% -32.8% Stops per Vehicle % -72.6% -31.4% Total Travel Time (hours) % -39.1% -11.3% Total Vehicles Entering Intersection 11,709 11,784 11, % -1.4% -2.1% Phase 2 Continuous Flow Intersections 51

72 Exhibit 4-16 provides a comparison of the delay by movement for each of the intersection design alternatives at the parkway-parkway intersection. Examination of the information contained in Exhibit 4-16 indicates the following: The primary weakness of the MLT design is the delay associated with the left-turn movements. The through and right-turn movements operate at or below capacity. The CFI design provides significantly better traffic operations and lower delay for all movements in comparison to the MLT design. All movements on the CFI operate well below capacity. All movements in the SPUI operate below capacity except the northbound and southbound left-turn movements, which could be improved through a minor modification of the signal timing to provide more green time for these movements. However, the change in signal timing would increase the delay to the higher volume eastbound and westbound through movements and increase total delay in the intersection. In general, the overall level of delay with the CFI is very similar to that of the SPUI under the assumed traffic volumes. Phase 2 Continuous Flow Intersections 52

73 Exhibit 4-16 PARKWAY-PARKWAY INTERSECTION (Mile 6) DELAY PER MOVEMENT BY INTERSECTION DESIGN ALTERNATIVE (MLT Uniform Capacity Volumes) Parkway / Parkway Intersection Delay per Vehicle by MLT Configuration (Single U-Turn Lane) with MLT Uniform Capacity Volumes Parkway / Parkway (Mile 6) Intersection Delay per Vehicle by CFI Configuration with MLT Uniform Capacity Volumes Delay / Veh (sec) Total Intersection Delay = 223 Hours Intersection Delay / Vehicle = 69 Seconds Total Intersection Stops = 13,754 Intersection Stops / Vehicle = 1.17 Total Intersection Travel Time = 400 Hours Total Vehicles = 11, EBL EBT EBR WBL WBT WBR NBL NBT NBR SBL SBT SBR 407 RT lane UT lane Other Delay Hourly Flow Rate (vph) Delay / Veh (sec) Total Intersection Delay = 102 Hours Intersection Delay / Vehicle = 31 Seconds Total Intersection Stops = 5,530 Intersection Stops / Vehicle = 0.47 Total Intersection Travel Time = 275 Hours Total Vehicles = 11, , , , ,086 EBL EBT EBR WBL WBT WBR NBL NBT NBR SBL SBT SBR RT lane LT lane Other Delay Hourly Flow Rate (vph) Parkway / Parkway Intersection Delay per Vehicle by SPUI Configuration with MLT Uniform Capacity Volumes Delay / Veh (sec) , Total Intersection Delay = 94 Hours Intersection Delay / Vehicle = 29 Seconds Total Intersection Stops = 3,714 Intersection Stops / Vehicle = 0.32 Total Intersection Travel Time = 244 Hours Total Vehicles = 11, , EBL EBT EBR WBL WBT WBR NBL NBT NBR SBL SBT SBR 398 1, , Delay in Th lanes 460 Hourly Flow Rate (vph) Phase 2 Continuous Flow Intersections 53

74 CFI CAPACITY TRAFFIC VOLUME SCENARIO The CFI Capacity scenario represents a traffic condition where the intersections are generally operating at capacity for the through movements. Note that the MLT Capacity volumes were not increased at the parkway-parkway intersection (Mile 6) to estimate the CFI Capacity volumes because the parkway-parkway intersection through movement was judged to be operating at capacity under the MLT Capacity volumes. Instead, the westbound rightturn volume at Mile 6 was increased to provide additional northbound traffic to estimate the capacity of the parkway at Mile 7 and Mile 10. The maximum hourly approach volumes used in the CFI Capacity analysis (4,600 vehicles per hour) represents an average daily traffic between 92,000 and 108,000 vehicles per day for the main parkway assuming the peak-hour traffic is between 10 and 8.5 percent of the daily traffic. This result is considered conservative in that the capacities of the left-turn and right-turn lanes were not explored in this study, and these lanes operated under capacity with the assumed traffic volumes. Considering only the through movement at signalized intersections, the CFI design capacity is estimated to be 1,000 to 1,050 vehicles per hour per through lane, which is very similar to the results for the MLT design. This result is as might be expected given that the MLT and CFI designs have the same two phase signal timing. The CFI intersection should be expected to have a slightly higher capacity because each vehicle only travels through the intersection once compared to the MLT design where left-turns travel through the intersection twice. A more detailed comparison of the traffic operations of the CFI and MLT intersections under their respective assumed capacity volumes is provided in Exhibit Examination of the results provided in Exhibit 4-17 indicates the following: The CFI and MLT intersections show very similar levels of delay for the through movements under the capacity volume conditions assumed in this study. This should be expected given the similarities between the signal phasing and timing between the two designs. The delay to the right-turn movements in the CFI design is substantially lower than in the MLT design due to the free flow right-turn lanes required by the CFI. The levels of delay reported in Exhibit 4-17 indicate that the right-turn volumes assumed in this study did not approach the capacity of the free flow right-turn lanes in the CFI. The capacity of the each free flow right-turn lane should exceed the capacity of a through lane (1,000 to 1,050 vph), but would be less than the ideal saturation flow rate for a single lane (1,900 vph). It is estimated that the capacity of a single free flow right-turn lane would be approximately 1,400 to 1,600 vehicles per hour depending on various traffic chacteristics and assuming an appropriately designed right-turn lane storage length and merge lane. The delay to the left-turn movements is substantially lower in the CFI than in an MLT intersection. However, the delay to left-turns in the MLT design is not directly indicative of the capacity for left-turns for this intersection type, because of the way delay for this movement is accumulated from the simulation. The total delay to left-turns in the MLT design is the sum of the delay as a through movement, a U-turn, and as the subsequent right-turn, plus the additional time spent traveling to and from the U-turn location. The most limiting factor to left-turn capacity in the MLT design is the capacity of the U-turn location, which was not evaluated in either this or the previous study. With the traffic Phase 2 Continuous Flow Intersections 54

75 volumes, the number of lanes, and signal timing assumed for this and the previous study, the queues in the left-turn lanes generally cleared each signal cycle for the MLT intersections. The left-turn movement delay in a CFI intersection is the sum of the delays at three signalized locations. Therefore, the total left-turn delay is not a direct indication of the capacity of any one of the three locations. For the simulation, every effort was made to coordinate the traffic signals at all three locations to allow the left-turn traffic to progress though the entire intersection while stopping for only the first signal encountered by the left-turn movement. The delay levels displayed in Exhibit 4-17 for the left-turn movement do not suggest that the assumed left-turn volumes reached or exceeded the capacity for the left-turn movements in these intersections. The free flow right-turn lane is a required feature of the CFI intersection, providing a significant traffic operations advantage to the CFI design that is absent in the MLT design. This feature could be added to the MLT design on each leg of an intersection where the associated U-turn opportunity is present. The application of free flow rightturn lanes would benefit traffic operations in an MLT design in two ways; by reducing delay to the right-turn movement, and by also reducing delay to the associated left-turn movement. This could significantly reduce the delay to MLT left-turn movements at locations where the traffic signal coordination cannot be established to eliminate the right-turn delay to the left-turn movement. Phase 2 Continuous Flow Intersections 55

76 Exhibit 4-17 COMPARISON OF DELAY BY MOVEMENT FOR THE MLT AND CFI DESIGNS WITH CAPACITY VOLUMES Delay / Veh (sec) Traffic Delay Per Vehicle Comparison Of CFI Configuration with CFI Capacity Volumes and MLT Configuration with MLT Capacity Volumes Through s EBT WBT NBT SBT EBT WBT NBT SBT Major Arterial Minor Arterial Cross Street and MLT Capacity Condition CFI Capacity Condition Delay / Veh (sec) Traffic Delay Per Vehicle Comparison Of CFI Configuration with CFI Capacity Volumes and MLT Configuration with MLT Capacity Volumes Right Turn s EBR WBR NBR SBR EBR WBR NBR SBR Major Arterial Cross Street and MLT Capacity Condition Minor Arterial CFI Capacity Condition Delay / Veh (sec) Traffic Delay Per Vehicle Comparison Of CFI Configuration with CFI Capacity Volumes and MLT Configuration with MLT Capacity Volumes Left Turn s EBL WBL NBL SBL EBL WBL NBL SBL Major Arterial 151 Cross Street and MLT Capacity Condition Minor Arterial CFI Capacity Condition Phase 2 Continuous Flow Intersections 56

77 CFI UNIFORM CAPACITY TRAFFIC VOLUME SCENARIO Comparison of the CFI Operations with CFI Capacity and Uniform Capacity Volumes A traffic volume scenario was developed to test the traffic operations of the CFI intersection under a condition where the traffic demand on all four intersection approaches approximated the through movement capacity of the intersection. The scenario was only tested at the parkwayparkway (Mile 6) intersection. The assumed traffic volumes for this scenario were provided earlier in Exhibit 3-5. An analysis was conducted to determine how the CFI intersection operated with uniform traffic volumes in comparison to the CFI Capacity volumes where the volumes were higher along the main north-south parkway. A table comparing the CFI operations under both traffic volume scenarios is provided in Exhibit In addition, the delay per vehicle by individual movement for each of the volume scenarios is provided in Exhibit These results indicate the following: The uniform capacity volumes resulted in a 3.5 percent increase in the total volume entering the intersection. Total delay increased by 20 percent and delay per vehicle increased by 17 percent with the uniform capacity volumes. Total stops increased by 13 percent and stops per vehicle increased by 9 percent. Total travel time in the intersection increased by 19 percent. The application of the uniform volumes had a significant negative impact on the through volume capacity of the intersection design in comparison to the use of the CFI Capacity volumes, which is to be expected. However, the uniform capacity of the CFI resulted in a total volume of vehicles entering the intersection that was 23 percent greater than that for the uniform capacity of the MLT design (14,364 versus 11,709 from Exhibit 4-15). This suggests that the CFI design should be expected to accommodate a significantly higher overall demand volume than the MLT design tested here. Phase 2 Continuous Flow Intersections 57

78 Exhibit 4-18 CFI CAPACITY TRAFFIC OPERATIONS CFI Summary Statistics Volumes Comparison Performance Measure CFI Parkway Capacity Volumes CFI Uniform Capacity Volumes Parkway Capacity to Uniform Capacity % Diff. Total Delay (hours) % Delay per Vehicle (seconds) % Total Stops 15,035 16, % Stops per Vehicle % Total Travel Time (hours) Total Vehicles Entering Network % 13,872 14, % Phase 2 Continuous Flow Intersections 58

79 Exhibit 4-19 CFI DELAY BY MOVEMENT FOR CFI CAPACITY AND UNIFORM CAPACITY VOLUMES Parkway / Parkway (Mile 6) Intersection Delay per Vehicle by CFI Configuration with CFI Capacity Volumes Delay / Veh (sec) Total Delay = 218 Hours Delay / Vehicle = 57 Seconds Total Stops = 15,035 Stops / Vehicle = 1.08 Total Travel Time = 394 Hours Total Vehicles = 13, , , , ,424 EBL EBT EBR WBL WBT WBR NBL NBT NBR SBL SBT SBR RT Lane LT Lane Other Delay Hourly Flow Rate (vph) Parkway / Parkway (Mile 6) Intersection Delay per Vehicle by CFI Configuration with CFI Uniform Capacity Volumes Total Delay = 264 Hours Delay / Vehicle = 66 Seconds Total Stops = 16,933 Stops / Vehicle = 1.18 Total Travel Time = 468 Hours Total Vehicles = 14,364 Delay / Veh (sec) , , , , RT lane LT lane Other Delay Hourly Flow Rate (vph) 0 EBL EBT EBR WBL WBT WBR NBL NBT NBR SBL SBT SBR Phase 2 Continuous Flow Intersections 59

80 Comparison of the CFI and SPUI Operations with CFI Uniform Capacity Volumes The SPUI design at the parkway-parkway (Mile 6) intersection was simulated using the CFI Uniform Capacity volumes for a comparison of traffic operations to the CFI intersection design. The comparison of the overall results is provided in Exhibit 4-20, and the comparison of the delay per vehicle by individual movement is provided in Exhibit The results indicate: The SPUI provides significantly better overall traffic operations under the assumed volumes than the CFI. The SPUI reduced delay by 42 percent, stops by 58 percent, and total travel time by 29 percent in comparison to the CFI. The SPUI provides much lower delay to the northbound and southbound through movements, as would be expected since these movements are free flow, but the SPUI also provides lower delay to the eastbound and westbound through movements. The CFI provides slightly lower levels of delay per vehicle for the left-turn movements. The CFI also provides lower levels of delay for the right-turn movements. Exhibit 4-20 CFI AND SPUI TRAFFIC OPERATIONS WITH CFI UNIFORM CAPACITY TRAFFIC VOLUMES CFI Uniform Capacity Intersection Type Comparison Performance Measure CFI SPUI CFI to SPUI % Diff. Total Delay (hours) % Delay per Vehicle (seconds) % Total Stops 16,933 7, % Stops per Vehicle % Total Travel Time (hours) Total Vehicles Entering Network % 14,364 14, % Phase 2 Continuous Flow Intersections 60

81 Exhibit 4-21 DELAY PER VEHICLE BY MOVEMENT FOR THE CFI AND SPUI WITH CFI UNIFORM CAPACITY VOLUMES Parkway / Parkway (Mile 6) Intersection Delay per Vehicle by CFI Configuration with CFI Uniform Capacity Volumes Total Delay = 264 Hours Delay / Vehicle = 66 Seconds Total Stops = 16,933 Stops / Vehicle = 1.18 Total Travel Time = 468 Hours Total Vehicles = 14, Delay / Veh (sec) , , , , RT lane LT lane Other Delay Hourly Flow Rate (vph) 0 EBL EBT EBR WBL WBT WBR NBL NBT NBR SBL SBT SBR Parkway / Parkway (Mile 6) Intersection Delay per Vehicle by SPUI Configuration with CFI Uniform Capacity Volumes Total Delay = 151 Hours Delay / Vehicle = 39 Seconds Total Stops = 7,039 Stops / Vehicle = 0.50 Total Travel Time = 333 Hours Total Vehicles = 14, Delay / Veh (sec) , , , , Hourly Flow Rate (vph) EBL EBT EBR WBL WBT WBR NBL NBT NBR SBL SBT SBR Phase 2 Continuous Flow Intersections 61

82 RIGHT-OF-WAY CONSIDERATIONS A brief planning level evaluation of right-of-way (ROW) considerations for the CFI was conducted as part of this study. Planning level of estimates of the ROW requirements for the MLT corridor and the SPUI application were developed in the previous study. The primary source of ROW requirements was the Maricopa County Department of Transportation Roadway Design Manual, Revised Exhibit 4-22 provides a MCDOT typical section for an urban principal arterial six-lane divided roadway. This typical roadway section includes a 14-foot raised median and a minimum 130- foot ROW. An estimate of the minimum ROW requirements for an eight-lane parkway with a CFI parkway-parkway, parkway-major arterial, and parkway-minor arterial intersection was obtained by adjusting the median width, assuming a width for each additional traffic lane, and assuming a separation distance between the left-turn, right-turn, and opposite direction through lanes in the intersection. The following assumptions were used: Eight lane divided parkway between intersections. Median width at the intersection = 6 feet. Each additional lane width = 12 feet. Separation between left-turn lanes and opposing through lanes (at the intersection) = 6 feet. Separation between the left-turn lanes and the opposing free flow right-turn lane (at the intersection) = 6 feet. The parkway has the same number of lanes, and therefore the same cross section at each intersection type. The assumptions above, combined with the information presented in Exhibit 4-22 provide a minimum ROW width of 201 feet assuming dual left-turn lanes and a single free flow right-turn lane on each approach. The results of the previous study indicated an approximate ROW width of 200 feet midblock on the eight-lane MLT parkway with a 225 foot width at major intersections. Therefore, the CFI ROW requirements appear consistent with the ROW required for the MLT parkway concept in general. Phase 2 Continuous Flow Intersections 62

83 Exhibit 4-22 MCDOT TYPICAL SECTION FOR AN URBAN PRINCIPAL ARTERIAL ROADWAY Source: MCDOT Roadway Design Manual, Revised CFI ACCESS CONSIDERATIONS A significant consideration with the application of a CFI is access to the properties in each of the intersection quadrants. The location of the left-turn lanes and the free flow right-turn lanes restricts access within the limits of these lanes. An alternative that has been used is to provide access to properties at the intersection via a frontage road system that connects to the main roadway down stream of the end of the free flow right-turn lane. This application is illustrated in the photograph of an actual CFI provided in Exhibit Placement of the access point and design of the frontage road system would depends on the individual development plan for the adjacent property. These access points should be right-in, right-out only. Phase 2 Continuous Flow Intersections 63

84 Exhibit 4-23 AERIAL VIEW OF A CFI WITH FRONTAGE ROADS FOR ACCESS TO ADJACENT PROPERTIES Frontage road Frontage road Source: ABMB Engineers, Inc., Baton Rouge, LA. Reproduced with permission. Phase 2 Continuous Flow Intersections 64

85 5. SUMMARY OF RESULTS AND CONCLUSIONS The following results and conclusions are based on the technical analysis conducted in this study. It should be noted that the results and conclusions based on the traffic simulation conducted for this study may be limited to the range of conditions applied and assumptions made in conducting the analysis, and cannot necessarily be generalized to all possible combinations of conditions. The capacity of an eight-lane divided roadway consisting of CFI intersections is estimated to be between 92,000 and 108,000 vehicles per day (vpd) assuming that the peak-hour volume is between 10 and 8.5 percent of the daily traffic. This result is considered somewhat conservative in that the capacities of the left-turn and right-turn lanes were not explored in this study, and these lanes operated under capacity with the assumed traffic volumes. Considering only the through movement at signalized intersections, the CFI design capacity is estimated to be approximately 1,000 to 1,050 vehicles per hour per through lane, which is slightly higher than the estimated 975 to 1,025 vphpl for the MLT design. The primary difference in through movement capacity of the CFI and the MLT intersection is that with the MLT intersection the left-turn movement volume must pass through the intersection, reducing the capacity for through traffic. The capacity of a CFI is estimated to be between 45 and 55 percent greater than the capacity of a conventional intersection design with multi-phase signal timing. The aggregate intersection performance measures indicate that the CFI reduced delay by 46 percent, reduced stops by 50 percent, and reduced travel time by 46 percent in comparison to the conventional intersection design. At the parkway-parkway (Mile 6) intersection the CFI reduced delay by 59 percent, stops by 69 percent, and travel time by 59 percent in comparison to the conventional intersection design. The CFI is similar to the MLT in terms of the treatment of the through traffic movement at the intersection. Both intersection designs provide simple two-phase signal timing for the through movement, significantly increasing through movement capacity in comparison to the multi-phase signal timing used with conventional intersection design. The primary reduction in delay, stops, and travel time with the CFI design is in the leftturn and right-turn movements in comparison to the MLT design. In the CFI, the leftturns only pass through the intersection configuration once as opposed to the MLT design where they pass through the intersection twice. The free flow right-turn lanes also provide significant advantage to the CFI design. A summary comparison of the performance metrics for the MLT intersection design and the CFI is provided in Exhibit 5-1 for each analysis scenario. These results indicate that the full CFI (left-turn traffic removed upstream of the intersection on each approach) provided significant reductions in delay, stops, and travel time in comparison to the MLT across all analysis scenarios and for each of the intersection types evaluated. The minimum reduction in delay for a full CFI in comparison to the MLT design was 19 percent, with a maximum of 55 percent. Free flow right-turn lanes could be included in the MLT design on each approach that provides the median U-turn movement for left-turns, but this feature was not tested as part of this study. The use of free flow right-turn lanes with the MLT design would Phase 2 Continuous Flow Intersections 65

86 reduce delay to both the right-turn movement and the associated left-turn movement, and could significantly improve the traffic operations of the MLT design. The CFI generated better performance metrics than the SPUI for the Base Capacity volumes (see Exhibit 5-2), which were the lowest volumes tested. The SPUI generated significantly better performance metrics than the CFI with the MLT Capacity and CFI Uniform Capacity volumes. The SPUI and the CFI provided comparable levels of delay with the MLT Uniform Capacity Volumes. The traffic operations performance improvements for the SPUI were as high as a 58 percent reduction in delay, a 69 percent reduction in stops, and a 39 percent reduction in travel time in comparison to the CFI design. The primary benefit of the SPUI design is the free flow operation of the grade separation for the major through movement of traffic on the parkway in the analysis. While it may be possible for the CFI to provide traffic operations characteristics similar to the SPUI at lower volume levels, this was not the case for the higher traffic volumes that approximated the capacity of the CFI intersection used in this study. A potential advantage of the CFI design over the SPUI is the ability of the CFI to accommodate high volumes of left-turn and right-turn movements with lower levels of delay than the SPUI. A disadvantage of the CFI design is the issue of providing access to adjacent properties on the corners of the intersection. The location of the left-turn lanes and the free flow right-turn lanes basically eliminates access within the limits of these lanes. Direct access should not be provided from the free flow right-turn lanes because of safety concerns. An alternative that has been used is to provide access to properties at the intersection via a frontage road system that connects to the main roadway down stream of the end of the free flow right-turn lane. Placement of the access point and design of the frontage road system would very much depend on the individual development plan for the adjacent property. These access points should be right-in, right-out only. The minimum right-of-way for an eight-lane divided parkway with CFI intersections consisting of dual left-turn lanes and a single free flow right-turn lane on each approach would be approximately 201 feet. This ROW is consistent with the estimated requirements for an MLT intersection corridor of approximately 200 feet, and is 24 feet less than the estimated 225 feet required for an MLT design at the parkway-parkway intersection. Phase 2 Continuous Flow Intersections 66

87 Exhibit 5-1 COMPARISON OF CFI TO MLT DESIGN BY ANALYSIS SCENARIO Volume Scenario Base Capacity MLT Capacity MLT Uniform Performance Metric CFI Percent Difference Analysis Travel Location Delay Stops Time Aggregate for 3 Intersections Parkway-Parkway Intersection (Mile 6) Parkway-Major Arterial Intersection (Mile 10) Parkway-Minor Arterial 1 Intersection (Mile 7) Aggregate for 3 Intersections Parkway-Parkway Intersection (Mile 6) Parkway-Major Arterial Intersection (Mile 10) Parkway-Minor Arterial Intersection (Mile 7) Parkway-Parkway Intersection (Mile 6) Capacity 1. Only a partial CFI was simulated at this location for this volume scenario, that is, the left-turns were removed from the main intersection for only the two approaches on the main parkway. Exhibit 5-2 COMPARISON OF SPUI TO CFI DESIGN BY ANALYSIS SCENARIO Volume Scenario Base Capacity MLT Capacity MLT Uniform Capacity CFI Uniform Capacity Performance Metric SPUI Percent Difference Analysis Travel Location Delay Stops Time Parkway-Parkway Intersection (Mile 6) Parkway-Parkway Intersection (Mile 6) Parkway-Parkway Intersection (Mile 6) Parkway-Parkway Intersection (Mile 6) Phase 2 Continuous Flow Intersections 67

88

89 Preparedby: 8710NorthThornydaleRoad,Suite140 Tucson,Arizona85742 phone fax