1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 An optimal enforcement strategy for bus lanes in Santiago de Chile Christopher Bucknell Email: cpbuckne@uc.cl Sebastián Tamblay Email: sjtambla@uc.cl Jaime Moya Email: jfmoya@uc.cl Juan Carlos Muñoz Departamento de Ingeniería de Transporte y Logística, Pontificia Universidad Católica de Chile Email: jcm@ing.puc.cl Matías Navarro Email: mns@ing.puc.cl Alejandro Schmidt Email: asg@ing.puc.cl Antonio Gschwender Gerencia de Desarrollo, Directorio de Transporte Público Metropolitano Email: antonioe.gschwender@dtpm.gob.cl 1
31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 Introduction Commercial speed is one of the key attributes of a public transport system, impacting not only users experience but also its total cost and efficiency (if travelling faster, the same transport capacity can be offered with less vehicles). In this line, transit travel time reductions can be achieved through preferential treatment for buses over other vehicles on streets, acknowledging public transport greater efficiency, occupancy and fewer negative externalities (Vuchic, 2005). One of the main bus priority tools are bus-only streets or bus lanes. In both cases, other vehicles may be allowed to enter these bus streets or lanes in order to access certain destinations or parking spaces, and cameras are usually deployed to enforce a set of given rules to access these reserved areas. However, in many developing countries monetary resources are highly limited and not all bus lanes have cameras installed. Therefore, part of this task is undertaken with enforcement personnel through site visits. In the case of Transantiago (the public transportation system of Santiago, Chile) bus lanes are enforced through both cameras and enforcement personnel. Bus lanes without cameras are highly breached when no enforcement is done. Visits to bus lanes are done periodically, without having a clear strategy in order to obtain the greatest benefit for users. Thus, there is a need for authorities to (i) identify bus lanes with the worst level-of-service, (ii) define an enforcement strategy to improve these lanes and (iii) the creation of a tool that systematically and automatically applies this enforcement strategy through the use of up-to-date data. This paper proposes an enforcement strategy with the objective of increasing the level-of-service of bus users, using available Automatic Vehicle Location (AVL) data. Then, after data processing and application of the proposed methodology, the strategy is applied to Transantiago. Finally, results are shown, presenting both the optimal allocation of enforcement teams and the consequential increase in buses commercial speed. Methodology and data Recent developments in Transantiago have allowed access to better transit information. Specifically, AVL and smartcard information have been processed into two main outputs: reconstruction of trips based on smartcards used in buses and metro (Munizaga & Palma, 2012; Munizaga et al., 2014), and bus speed 2
58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 profiles (Cortés et al., 2011; DTPM, 2015) which can be aggregated to obtain average times and speeds of different bus lanes. Once speed data is processed, the methodology consists in three steps: a) Identify bus lanes in which enforcement is necessary Bus lanes prone to be enforced are going to be those who currently do not have cameras installed. Travel time measurements plus flow of buses will define a proxy to indicate the severity of the problem. b) Definition of an enforcement strategy Until now, bus lanes are enforced only at one point of the lane each time they are visited (independent of the length of the bus lane). In practice, cars who infringed the bus lane change lanes when they spot the enforcement team, and then incorporated themselves once again into the bus lane downstream. Additionally, the assignment of teams to different bus lanes is currently made without taking into account buses commercial speeds and potential benefits in travel time savings. We propose a different strategy, where bus lanes are enforced with teams strategically located along the complete bus lane. Since enforcement teams are limited, the number of visits to each bus lane is going to be scheduled proportional to an estimate of travel time savings per enforcement team. Later on, visits will be scheduled proportionally to the actual time savings per enforcement team. c) Strategy deployment In order to determine the optimal visit scheduling for each site, an optimization problem was formulated within a fixed-time window for planning purposes. This problem was written in AMPL and solved using CPLEX (Fourer et al., 1993). Results The result of the proposed methodology is an optimal enforcement strategy for bus lanes, where teams are assigned to the points where they will be more effective in increasing buses speeds, but keeping in mind that no problematic bus lane is left behind (instead, less critical lanes are visited less often). This strategy is currently under development and has not been applied on the streets yet. The set of bus lanes to enforce is already identified (along with the specific locations for the enforcement teams in each one of them). Likewise, the optimization problem is already formulated and tested with positive results. 3
84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 We expect our first field results (i.e., application of the enforcement strategy) during May 2016. With these first results, we will analyze actual speed improvements and study potential weight modifications for the scheduling problem, as stated above. Conclusions and perspectives We presented a generic methodology that allows an optimal enforcement visits scheduling for bus lanes, considering the potential benefits in travel time savings in order to allocate enforcement personnel to where it can have larger impacts. The methodology is applied to Transantiago, where enforcement scheduling is currently done with a strategy not directly related to buses commercial speeds. We expect our results to show that significant improvements in buses travel times can be achieved through simple strategies, effectively exploiting automated data and ensuring more efficient use of available enforcement personnel. Future research should include actual passenger flows in the buses using the different lanes, as the implemented strategy only considers bus flows. Some information regarding passengers load profiles can be obtained from the automated information available for Transantiago. However, there is a high and heterogenous bus fare evasion in the city, reaching level over 25% (DTPM, 2015), meaning that smartcard trip information is incomplete. References Cortés, C., Gibson, J., Gschwender, A., Munizaga, M., & Zúñiga, M. (2011). Commercial bus speed diagnosis based on GPS-monitored data. Transportation Research Part C: Emerging Technologies, 19(4), 695-707. DTPM (2015) Informe de Gestión 2014. Directorio de Transporte Público Metropolitano. Retrieved (2016, March 14) from: http://www.dtpm.gob.cl/archivos/informe_gestion-2014_vfinal Fourer, R., Gay, D. M., & Kernighan, B. W. (1993). AMPL: A Modeling Language for Mathematical Programming. The Scientific Press, South San Francisco, California. Munizaga, M., & Palma, C. (2012). Estimation of a disaggregate multimodal public transport Origin Destination matrix from passive smartcard data from Santiago, Chile. Transportation Research Part C: Emerging Technologies, 24, 9-18. 4
111 112 113 114 Munizaga, M., Devillaine, F., Navarrete, C., & Silva, D. (2014). Validating travel behavior estimated from smartcard data. Transportation Research Part C: Emerging Technologies, 44, 70-79. Vuchic, V. (2005). Urban Transit: Operations, Planning and Economics. John Wiley & Sons, Inc. Hoboken, New Jersey. 5