Urban public transport optimization by bus ways: a neural network-based methodology

Similar documents
The safe ships trajectory in a restricted area

Strategic Decision Making in Portfolio Management with Goal Programming Model

Transit Priority Strategies for Multiple Routes Under Headway-Based Operations

Development of Urban Public Transit Network Structure Integrating Multi-Class Public Transit Lines and Transfer Hubs

Real-time Stochastic Evacuation Models for Decision Support in Actual Emergencies

Using Rates of Change to Create a Graphical Model. LEARN ABOUT the Math. Create a speed versus time graph for Steve s walk to work.

A Probabilistic Approach to Worst Case Scenarios

The Measuring System for Estimation of Power of Wind Flow Generated by Train Movement and Its Experimental Testing

Examining the limitations for visual anglecar following models

Economics 487. Homework #4 Solution Key Portfolio Calculations and the Markowitz Algorithm

Interpreting Sinusoidal Functions

SURFACE PAVEMENT CHARACTERISTICS AND ACCIDENT RATE

Capacity Utilization Metrics Revisited: Delay Weighting vs Demand Weighting. Mark Hansen Chieh-Yu Hsiao University of California, Berkeley 01/29/04

Application of System Dynamics in Car-following Models

Monte Carlo simulation modelling of aircraft dispatch with known faults

AP Physics 1 Per. Unit 2 Homework. s av

Paul M. Sommers David U. Cha And Daniel P. Glatt. March 2010 MIDDLEBURY COLLEGE ECONOMICS DISCUSSION PAPER NO

The design of courier transportation networks with a nonlinear zero-one programming model

KINEMATICS IN ONE DIMENSION

Time & Distance SAKSHI If an object travels the same distance (D) with two different speeds S 1 taking different times t 1

3.00 m. 8. At La Ronde, the free-fall ride called the Orbit" causes a 60.0 kg person to accelerate at a rate of 9.81 m/s 2 down.

Bootstrapping Multilayer Neural Networks for Portfolio Construction

As time goes by - Using time series based decision tree induction to analyze the behaviour of opponent players

Bill Turnblad, Community Development Director City of Stillwater Leif Garnass, PE, PTOE, Senior Associate Joe DeVore, Traffic Engineer

Performance Attribution for Equity Portfolios

8/31/11. the distance it travelled. The slope of the tangent to a curve in the position vs time graph for a particles motion gives:

Morningstar Investor Return

COMPARING SIMULATED ROAD SAFETY PERFORMANCE TO OBSERVED CRASH FREQUENCY AT SIGNALIZED INTERSECTIONS

Dual Boost High Performances Power Factor Correction (PFC)

A Liability Tracking Portfolio for Pension Fund Management

Revisiting the Growth of Hong Kong, Singapore, South Korea, and Taiwan, From the Perspective of a Neoclassical Model

What the Puck? an exploration of Two-Dimensional collisions

ANALYSIS OF RELIABILITY, MAINTENANCE AND RISK BASED INSPECTION OF PRESSURE SAFETY VALVES

CALCULATION OF EXPECTED SLIDING DISTANCE OF BREAKWATER CAISSON CONSIDERING VARIABILITY IN WAVE DIRECTION

DYNAMIC portfolio optimization is one of the important

Evaluating the Performance of Forecasting Models for Portfolio Allocation Purposes with Generalized GRACH Method

EXAMINING THE FEASIBILITY OF PAIRED CLOSELY-SPACED PARALLEL APPROACHES

The Current Account as A Dynamic Portfolio Choice Problem

Proportional Reasoning

A Study on the Powering Performance of Multi-Axes Propulsion Ships with Wing Pods

AGENDA REQUEST. September 7, 2010 Timothy Litchet

Methods for Estimating Term Structure of Interest Rates

2. JOMON WARE ROPE STYLES

Instruction Manual. Rugged PCB type. 1 Terminal Block. 2 Function. 3 Series Operation and Parallel Operation. 4 Assembling and Installation Method

The t-test. What We Will Cover in This Section. A Research Situation

Evaluating Portfolio Policies: A Duality Approach

Overview. Do white-tailed tailed and mule deer compete? Ecological Definitions (Birch 1957): Mule and white-tailed tailed deer potentially compete.

DRAFT FINAL MEMORANDUM

Proceedings of the ASME 28th International Conference on Ocean, Offshore and Arctic Engineering OMAE2009 May 31 - June 5, 2009, Honolulu, Hawaii

An Alternative Mathematical Model for Oxygen Transfer Evaluation in Clean Water

Reliability Design Technology for Power Semiconductor Modules

Lifecycle Funds. T. Rowe Price Target Retirement Fund. Lifecycle Asset Allocation

Evaluation of a car-following model using systems dynamics

2017 MCM/ICM Merging Area Designing Model for A Highway Toll Plaza Summary Sheet

Asset Allocation with Higher Order Moments and Factor Models

Name Class Date. Step 2: Rearrange the acceleration equation to solve for final speed. a v final v initial v. final v initial v.

Flexible Seasonal Closures in the Northern Prawn Fishery

LEWA intellidrive. The mechatronic All-in-One pump system. intelligent flexible dynamic high precision. Foto: ratiopharm

Homework 2. is unbiased if. Y is consistent if. c. in real life you typically get to sample many times.

A computational model to assess the impact of policy measures on traffic safety in Flanders: Theoretical concepts and application.

FHWA/IN/JTRP-2009/12. Panagiotis Ch. Anastasopoulos Fred L. Mannering John E. Haddock

Market timing and statistical arbitrage: Which market timing opportunities arise from equity price busts coinciding with recessions?

Automatic air-main charging and pressure control system for compressed air supplies

Improving Measurement Uncertainty of Differential Pressures at High Line Pressures & the Potential Impact on the Global Economy & Environment.

KEY CONCEPTS AND PROCESS SKILLS. 1. An allele is one of the two or more forms of a gene present in a population. MATERIALS AND ADVANCE PREPARATION

Measuring Potential Output and Output Gap and Macroeconomic Policy: The Case of Kenya

Constructing Absolute Return Funds with ETFs: A Dynamic Risk-Budgeting Approach. July 2008

QUANTITATIVE FINANCE RESEARCH CENTRE. Optimal Time Series Momentum QUANTITATIVE FINANCE RESEARCH CENTRE QUANTITATIVE F INANCE RESEARCH CENTRE

Market Timing with GEYR in Emerging Stock Market: The Evidence from Stock Exchange of Thailand

Portfolio Strategies Based on Analysts Consensus

PRESSURE SENSOR TECHNICAL GUIDE INTRODUCTION FEATURES OF ELECTRIC PRESSURE SENSOR. Photoelectric. Sensor. Proximity Sensor. Inductive. Sensor.

ScienceDirect. Cycling Power Optimization System Using Link Models of Lower Limbs with Cleat-Shaped Biaxial Load Cells

ITG Dynamic Daily Risk Model for Europe

Zelio Control Measurement Relays RM4L Liquid Level Relays

Explore Graphs of Linear Relations. 1. a) Use a method of your choice to determine how much water will be needed each day of a seven-day cruise.

Modeling the Repair Process ofa Power Distribution System

CMA DiRECtions for ADMinistRAtion GRADE 6. California Modified Assessment. test Examiner and Proctor Responsibilities

AN ANALYSIS OF THE ECONOMIC EFFECT OF A ROAD DIET IN ELIZABETHTOWN AND GEORGETOWN, KENTUCKY

Keywords: overfishing, voluntary vessel buy back programs, backward bending supply curve, offshore fisheries in Taiwan

Performance Optimization of Markov Models in Simulating Computer Networks

Chapter : Linear Motion 1

Avoiding Component Failure in Industrial Refrigeration Systems

Semi-Fixed-Priority Scheduling: New Priority Assignment Policy for Practical Imprecise Computation

A Stable Money Demand: Looking for the Right Monetary Aggregate

Dynamics of market correlations: Taxonomy and portfolio analysis

The APT with Lagged, Value-at-Risk and Asset Allocations by Using Econometric Approach

Stock Return Expectations in the Credit Market

Nozzle valve system for the variable atomization of very small liquid volumes

CHARACTERIZATION AND MODELING OF A PROPORTIONAL VALVE FOR CONTROL SYNTHESIS

AMURE PUBLICATIONS. Working Papers Series

SPECIAL WIRE ROPES The Value Line

COUPLING VOF/PLIC AND EMBEDDING METHOD FOR SIMULATING WAVE BREAKING ON A SLOPING BEACH

The credit portfolio management by the econometric models: A theoretical analysis

WHO RIDE THE HIGH SPEED RAIL IN THE UNITED STATES THE ACELA EXPRESS CASE STUDY

CALCULATORS: Casio: ClassPad 300 ClassPad 300 Plus ClassPad Manager TI: TI-89, TI-89 Titanium Voyage 200. The Casio ClassPad 300

A BETTER WAY THRU ROCKAWAY

3. The amount to which $1,000 will grow in 5 years at a 6 percent annual interest rate compounded annually is

WELCOME! PURPOSE OF WORKSHOP

What is a Practical (ASTM C 618) SAI--Strength Activity Index for Fly Ashes that can be used to Proportion Concretes Containing Fly Ash?

Nozzle valve system for the variable atomization of very small liquid volumes

Transcription:

Urban Transpor XIII: Urban Transpor and he Environmen in he 21s Cenury 347 Urban public ranspor opimizaion by bus ways: a neural nework-based mehodology M. Migliore & M. Caalano Deparmen of Transporaion Engineering, Palermo Universiy, Ialy Absrac This paper describes an approach for planning he inroducion of bus lanes ino he urban road nework, which has been applied o he urban area of Palermo. The proposed modelling ool adops a muli-agen objecive funcion expressing he rade-off beween he ineress of diverse sakeholders: he generalized ranspor cos for car drivers and he ravel ime for public ranspor users. The reacion of car raffic o a cerain planning scenario has been simulaed by he DUE assignmen echnique and he posiive impac of he modal shif on he objecive funcion has been ackled by aaching a suiable weigh o he ime saving for bus passengers. The rise in he bus ravel speed, owing o he bus lane soluion, has been prediced for a se of urban roads by a neural nework, so as o ake ino accoun many quaniaive and qualiaive road aribues. The opimal locaion paern of bus ways has been searched by a greedy heurisic ha hrough a sep-by-sep sraegy builds he problem soluion by keeping, a each sage, he bes alernaive. Keywords: urban public ranspor, bus way design, welfare funcion opimizaion, neural nework, greedy heurisic. 1 Inroducion Bus lanes represen an effecive ool for buses o bypass he congesion a criical secions of he urban road nework, especially during rush hours. Furhermore, hey consis in a ranspor policy acion ha can be easily implemened. In deail, bus lanes imply he following advanages: A rise in he bus ravel speed, generally speaking, from 10 Km/h o 13-15 Km/h. The puncualiy of he public ranspor service. WIT Transacions on The Buil Environmen, Vol 96, www.wipress.com, ISSN 1743-3509 (on-line) doi:10.2495/ut070331

348 Urban Transpor XIII: Urban Transpor and he Environmen in he 21s Cenury Comfor and safey improvemens, hanks o he reducion in acceleraions and deceleraions (sop and go driving). A decrease in he environmenal impac and energy consumpions. 2 The mobiliy sysem of Palermo The General Urban Transpor Plan of Palermo esimaes he inernal movemens a 960000 ravellers per day; he 81% of such mobiliy demand uses he car, whereas he residual par chooses he public ranspor, which highlighs a criical siuaion for Palermo, since, generally speaking, wih reference o he public ranspor share, saisics show a 18%-30% range in he case of small/mediumsized owns and values higher han 50% in he big own case. Moreover, he average ravel speed on he urban road nework is 18 km/h, while, inside he cenre, he ravel speed can be over he 7 km/h level, during rush hours. Oher alarming daa regard he number of accidens per annum wihin he urban area of Palermo ha comes o abou 2000. In order o reverse his rend, he local adminisraors have recenly launched he Inegraed Plan of Transpor, for a more effecive public ranspor sysem, which suggess he bus lane sysem enhancemen (Palermo Ciy Council [1]). 3 Problem definiion and model framework The research aciviy presened has consised in developing a quaniaive mehodology for planning he inroducion of bus ways ino he urban road nework. The resuling approach has been applied o Palermo ciy: he presen bus lane sysem has been compared wih he model-based one. In paricular, o idenify he sudy area wihin Palermo and o selec he roads for he bus lane design model, he auhors have considered parameers concerning he public ranspor level of service, number of lines and frequency, and geomeric parameers such as he road widh. The considered planning problem has been solved using an objecive funcion represening he rade-off beween he ineress of privae ranspor users and public ranspor ones: ( C, T T ) O. F. = f, (1) car where, C car : rise in he ranspor moneary cos for car users; T car : rise in he ravel ime for car users; T bus : reducion in he ransi ime for bus users. The choice behaviour of car drivers has been simulaed hrough he DUE assignmen mehod (Cascea [2]). car bus WIT Transacions on The Buil Environmen, Vol 96, www.wipress.com, ISSN 1743-3509 (on-line)

Urban Transpor XIII: Urban Transpor and he Environmen in he 21s Cenury 349 The effec of bus ways on he bus ravel speed has been evaluaed by a neural nework (Bishop [3]), allowing for several road aribues. The auhors have also considered he posiive impac of he modal shif on he local communiy welfare by assigning a suiable weigh o he ime saving for bus passengers. The opimal se of bus ways has been deermined hrough ieraions of a greedy algorihm (Tadei and Della Croce [4]). 4 Soluion process 4.1 The inpu for he bus lane design model To se he poenial sies for bus lanes, he auhors have seleced hose roads respecing he following crieria: one regarding he widh of he road secion for buses; anoher one consising in he frequency of he public ranspor service. In Table 1, he reader can analyze a marix reporing he differen ranspor policy acions ha he auhors propose for various combinaions of he aforesaid crieria. Table 1: Transpor policy acions based on he public ranspor level of service and road parameers. Crieria Public ranspor Widh of he road secion Transpor policy acion frequency, F for buses, L * (meers) (buses per hour) F 21 L 5 Bus lane F 21 L < 5 Closed o raffic zone 12 F < 21 L 5 The bus lane soluion is o be weighed 12 F < 21 L < 5 No acion F < 12 No acion * I excludes he space for car parking. 4.2 The esimaion of he objecive funcion s componens As saed before, he opimal bus lane sysem for Palermo has been searched adoping he following muli-agen objecive funcion: ( C, T T ) O. F. = f, (2) car car bus WIT Transacions on The Buil Environmen, Vol 96, www.wipress.com, ISSN 1743-3509 (on-line)

350 Urban Transpor XIII: Urban Transpor and he Environmen in he 21s Cenury In deail: T bus = C car = [ euros per car - km cars - km] ; T car = [ euros per car - hour cars - hour] ; [ euros per bus passenger - hour bus passengers - hour] where, cars-km: variaion of he cars-km variable owing o he inroducion of bus ways. Such quaniy refers o he number of privae vehicles moving on he nework and he disances covered by hemselves. cars-hour: variaion of he cars-hour variable arising from a cerain bus lane paern. This variable measures he ransi ime spen by privae vehicles. bus passengers-hour: variaion of he passengers-hour variable arising from a cerain bus lane paern. This variable measures he ransi ime spen by bus users. 4.2.1 The deerminaion of bus passengers-hour To compue bus passengers-hour, wo elemens need o be evaluaed: he public ranspor ravel speed for he roads under analysis (namely, he poenial locaions for bus ways) and, for each of hese ones, he number of he public ranspor cusomers. In paricular, o predic he public ranspor ravel speed, he auhors have adoped a neural nework, whose archiecure is he mulilayer percepron one (Bishop, [3]). In deail, a hree layer srucure has been used: Inpu layer: wih a number of neurons equal o he one of he variables observed for a generic road secion. I ransfers inpu daa o he processing layer hrough linear mahemaical funcions. Hidden layer: he compuing core of he neural nework wih a number of neurons deermined by he pruning approach. I employs he anh ransfer funcion. Oupu layer: wih a neuron for esimaing he bus ravel speed. Such an arificial inelligence ool processes empirical daa regarding he following quaniaive and qualiaive road aribues: bus way availabiliy (Y/N); number of bus sops; number of juncions wih raffic lighs; number of side roads; number of pedesrian crossings; possibiliy of parking wih he line of march (Y/N); ransgression of he parking rules (Y/N); presence of shops (Y/N); raffic flow/road capaciy (La Franca e al [5]). On he basis of he above-lised variables, he urban roads under analysis have been subdivided ino homogeneous secions. Moreover, he aforesaid empirical road daa have been normalized, o be processed by he neural nework as dimensionless and comparable inpus. In order o calculae he number of bus passengers, he origin-desinaion marix represening he public ranspor demand of Palermo during he peak hour of a working day (2002) has been assigned o he corresponding graph by he. WIT Transacions on The Buil Environmen, Vol 96, www.wipress.com, ISSN 1743-3509 (on-line)

Urban Transpor XIII: Urban Transpor and he Environmen in he 21s Cenury 351 hyperpah minimum algorihm, aking ino accoun he rise of he bus ravel speed, due o he inroducion of bus lanes (Cascea [2]). The oupu of he described bus passengers-hour esimaion process can be analyzed in Tables 2 and 3: Table 2: The neural nework-based oupu. Road name Via Noarbarolo Viale Lazio Via Basile Via Dane Via Pone di Mare Secions Lengh (km) Bus speed wihou bus lane (Km/h) Bus speed wih bus lane (Km/h) 1 0.15 6.57 14.58 2 0.25 7.07 7.57 3 0.25 6.31 7.96 4 0.2 7.88 15.06 5 0.15 6.28 13.75 1 0.25 7.00 7.10 2 0.3 6.98 10.46 3 0.24 6.31 9.77 4 0.24 6.52 8.16 1 0.70 14.12 16.37 2 0.35 15.09 15.94 3 0.50 9.84 15.99 4 0.45 11.38 15.17 1 0.3 5.94 11.77 2 0.15 5.88 9.16 3 0.4 6.08 11.09 4 0.35 7.91 8.77 1 0.27 9.22 15.94 2 0.47 7.66 15.15 3 0.38 7.55 8.89 4 0.47 7.88 13.56 4.2.2 The deerminaion of cars-km and cars-hour To appraise cars-km and cars-hour, he origin-desinaion marix represening he privae ranspor demand of Palermo during he peak period of a working day (2002) has been assigned by he DUE approach o he corresponding graph, aking ino accoun he reducion in space for car movemens, due o he inroducion of bus lanes. 4.3 Compuaional resuls This secion shows he oucomes achieved by he greedy algorihm (Tadei and Della Croce [4]) applied o deermine he opimal se of bus ways for he public ranspor sysem of Palermo. WIT Transacions on The Buil Environmen, Vol 96, www.wipress.com, ISSN 1743-3509 (on-line)

352 Urban Transpor XIII: Urban Transpor and he Environmen in he 21s Cenury Table 3: The neural nework-based oupu. Road name Corso Albero Amedeo Corso Tukory Secions Lengh (km) Bus speed wihou bus lane (km/h) Bus speed wih bus lane (km/h) 1 0.2 6.44 13.64 2 0.4 9.86 15.81 1 0.2 11.72 12.20 2 0.2 13.51 12.54 3 0.2 9.51 15.74 4 0.15 6.74 14.52 Via Goehe 1 0.35 7.54 11.16 1 0.4 7.04 15.06 Via Duca della Verdura Via Marchese di Villabianca 2 0.2 6.36 14.02 1 0.1 7.15 14.84 2 0.15 6.04 9.09 3 0.25 5.95 10.02 4 0.15 7.07 13.39 5 0.15 9.70 13.80 6 0.1 10.40 15.95 The above-menioned heurisic consrucs he bus lane design problem soluion hrough a sep-by-sep opimizaion process and he following objecive funcion: W.F. = α ( bus passengers-hour cars-hour) β cars-km (3) where, W.F.: Welfare Funcion α: value of ime (2.6 euros per passenger-hour) β: privae ranspor cos (0.1 euros per car-km) The value of ime characerizing he mobiliy sysem of Palermo can be esimaed a 2 euros/hour. Such a figure has been muliplied by 1.3, so as o ake ino accoun, on one hand, he posiive environmenal impac of a more compeiive public ranspor service and, on he oher hand, he need of urning cars-hour ino car passengers-hour; in fac, he privae ranspor occupancy facor for Palermo can be appraised a 1.3 ravellers per vehicle (Palermo Ciy Council [1]). As a consequence: W.F. = (2 euros per hour) 1.3 ( bus passengers-hour cars-hour), (0.1 euros per car-km) cars-km. (4) The opimal se of bus ways has been buil by a sequenial approach: a each sage ( = 1, 2,, N), amongs he available roads wih a criical public WIT Transacions on The Buil Environmen, Vol 96, www.wipress.com, ISSN 1743-3509 (on-line)

Urban Transpor XIII: Urban Transpor and he Environmen in he 21s Cenury 353 Table 4: The comparison beween he pre-exisen bus way nework of Palermo and he model-based one. Road name The presen bus lane sysem of Palermo The model oupu Corso Tukory Via Dane (firs secion)* - Via Dane (second and hird secions) - Via Duca Della Verdura Via Goehe - Via Liberà* Via Maqueda* Via Marchese di Villabianca - Via Noarbarolo Via Oreo* - Via Pone di Mare - Via Roma* Viale Lazio - Viale Srasburgo* Via Erneso Basile - - Corso Albero Amedeo - : wih a bus lane; -:wihou a bus lane. *: roads wih a so high bus frequency ha hey have been assigned a bus way ou of he model. ranspor frequency (lying beween 12 and 21 buses/hour; see Table 1), he algorihm idenifies he one maximizing he Welfare Funcion by he inroducion of a bus lane. In formal erms: 1. = 1 Max W. F., subjec o x = 1 2. ( ) x x 3. If maxw. F. ( x ) 0 x 4. Else x = arg maxw. F. ( x ) x 5. = + 1 6. + 1 MaxW. F. x x x j= 1 2,, -1 k hen sop j k, subjec o xk = 1 and k j x x, j = 1, WIT Transacions on The Buil Environmen, Vol 96, www.wipress.com, ISSN 1743-3509 (on-line)

354 Urban Transpor XIII: Urban Transpor and he Environmen in he 21s Cenury a b Table 5: The opimal bus lane locaion a each sage of he greedy algorihm a. Greedy seps Via Noarbarolo Viale Lazio Via Basile Impac (euros) on he Welfare Funcion if a bus lane is placed in: Via Dane b Via Pone di Mare Corso A. Amedeo Via Goehe Corso Tukory Via Duca della Verdura Via Marchese di Villabianca 1 +48.48 +3.79-26.74 +39.67 +15.39-3.92 +15.77-4.67 +105.93 +57.91 2 +69.72 +25.38-1.49 +63.03 +15.66 +8.65 +16.05 +17.17 - +27.24 3 - +15.78 +3.51 +56.53 +30.87 +11.84 +19.34 +14.07 - +39.64 4 - +25.98-6.19 - +43.16-2.25 +23.05 +18.07 - +44.04 5 - +18.78-1.99 - +24.96-11.85 +14.97 +13.65 - - 6 - +22.58 +0.11 - - +9.05 +26.55 +13.77 - - 7 - +8.38-16.89 - - -6.25 - +9.97 - - 8 - +24.18-4.39 - - +5.85 - - - - 9 - - -5.29 - - -10.55 - - - - A every sep, he opimal soluion is wrien in bold face; - means ha he corresponding sree has been seleced as bes sie for a bus lane a a previous ieraion. I refers o he second secion and he hird one. WIT Transacions on The Buil Environmen, Vol 96, www.wipress.com, ISSN 1743-3509 (on-line)

Urban Transpor XIII: Urban Transpor and he Environmen in he 21s Cenury 355 7. 1 If max.. + j W F x x 0 or = N hen sop x j= 1 8. Else = + 1 j x arg maxw. F. x x x j= 1 9. Go o sep 5. where, x: vecor represening he locaion paern of bus lanes, whose generic elemen is x k belonging o he se of hose roads wih a criical public ranspor frequency (lying beween 12 and 21 buses/hour); x k :1 if road k is provided wih a bus way; 0 oherwise. As for he applicaion oupu, he reader can noice in Table 5 ha he greedy algorihm sops a he eighh ieraion, since a he ninh sep, each of he available alernaives (Via Erneso Basile and Corso Albero Amedeo) enails a negaive value of he objecive funcion. 5 Conclusions A clear proof of he described mehodology uiliy emerges from he oucome showed in Table 5. In fac, consider he Corso Albero Amedeo case (see Table 3): he public ranspor is characerized by an average demand of 440 passengers per hour and an average ravel speed of 8.15 km/h; hanks o a bus lane, he average speed level of 14.72 km/h can be gained, which means a remarkable 81% rise. Consequenly, he bus lane for Corso Albero Amedeo seems o be an opporune ool o improve he public ranspor performance; on he conrary, under an overall poin of view concerning he urban mobiliy sysem, he illusraed planning model excludes he bus way soluion for Corso Albero Amedeo. Furhermore, he comparison beween he presen bus lane sysem of he sudy area and he one resuling from he described mehodology proves ha he former ha is based on a qualiaive planning approach needs o be highly developed as i has been showed in his paper (see Table 4). References [1] Palermo Ciy Council, Final Repor Urban Transpor Plan, 2002. [2] Cascea, E., Transporaion Sysems Engineering: Theory and Mehods, Dordrech (The Neherlands): Kluwer Academic Publishers, 2001. [3] Bishop, C. M., Neural Nework for Paern Recogniion, Oxford Universiy Press, Oxford, 1995. [4] Tadei, R., Della Croce, F., Ricerca Operaiva e Oimizzazione, Esculapio, Bologna, 2001. WIT Transacions on The Buil Environmen, Vol 96, www.wipress.com, ISSN 1743-3509 (on-line)

356 Urban Transpor XIII: Urban Transpor and he Environmen in he 21s Cenury [5] La Franca, L., Migliore, M., Salvo, G., Carollo, F., The auomaic vehicle monioring o improve he urban public ranspor managemen, Proc. of he World Congress CODATU XI Towards More Aracive Urban Transporaion, Bucares, 2004. WIT Transacions on The Buil Environmen, Vol 96, www.wipress.com, ISSN 1743-3509 (on-line)