Optimal scheduling of public transport fleet at network level

This paper deals with the development of a strategic approach for optimizing the operation of public transport system that considers both user's objective and operator's objective. Passengers of public transport are assumed to seek a minimum wait time to conduct the trips, while on the other hand, operators are concerned with the efficient operation such as minimum fleet size. The average minimum wait time is to be achieved by creating an optimal despatching policy for each vehicle from the terminal. As for efficient operation the utilisation of vehicle should be maximised by having a minimum number of vehicles in operation. User's and operator's objectives are optimized within certain operational constraints such as vehicle capacity to maintain acceptable level of service. The i‐model is contructed in a bi‐level programming form in which the user's objective is minimized by dynamic programming and the operator's objective is minimized by various routing strategies. Furthermore, an algorithm and a contrived example are developed to solve and see the performance of the approach.     

Public transportation development and traffic accident prevention in Indonesia

Traffic accidents have long been known as an iceberg for comprehending the discrepancies of traffic management and entire transportation systems. Figures detailing traffic accidents in Indonesia, as is the case in many other countries, show significantly high numbers and severity levels; these types of totals are also evident in Jakarta, the highest-populated city in the country. While the common consensus recognizes that traffic accidents are the results of three different factor types, namely, human factors, vehicle factors, and external factors (including road conditions), human factors have the strongest influence—and figures on a worldwide scale corroborate that assertion. We, however, try to pinpoint the issues of non-human factors in light of increasing traffic accidents in Indonesia, where motorbike accidents account for the majority of incidents. We then consider three important pillars of action: the development of public transportation, improvement of the road ratio, and traffic management measures.     

Traffic flow consideration in design of freight distribution system

This study is part of a series of research projects on a distribution system we developed to deal with cases in a state-owned company. It concerns the design of the Public Service Obligation State-owned Company (PSO-SOC) distribution system. The intrinsic features of PSO-SOC are distributing strategic commodities and having subsidies within the cost function. Hence their distribution flow has to be secured under consideration of moving the commodities within road networks that have traffic flow dependency. This paper focuses on the solution of the proposed model which represents traffic flow dependency within a freight distribution network.The mathematical formulation takes the form of a Minimum Cost Multicommodity Flow (MCMF) problem. Traffic flow dependency is incorporated into the model by introducing a coefficient of speed, which is derived from the traffic assignment of ordinary traffic associated with the transportation of the type of freight under consideration The solution of the proposed model is formulated by Network Representation (NR), in which all of the components of the mathematical model are represented in the form of dummy links and nodes added to the original (physical) network. It is to be noted then, that the traffic flow on each road or link is represented by a link performance function (LPF), depicting traffic flow dependent travel time and consequent cost. The MCMF problem of NR is further solved by a Primal-Dual Algorithm.Finally, an illustrative example is exercised to show how the proposed step-wise solution works.     

Design of public transit network in urban area with elastic demand

This research focuses on an efficient design of transit network in urban areas. The system developed is used to create, analyze and optimize routes and frequencies of transit system in the network level. The analysis is based on elastic demand, so the shift of demand between modes in network due to different service level is of prime consideration. The developed system creates all feasible routes connecting all pairs of terminals in the network. Out of this vast pool of routes, a set of optimal routes is generated for a certain predetermined number that maintains connectivity of significant demand. Based on these generated routes, the system fulfils transportation demand by assigning demand that considers path and route choices for non-transit users and transit users. Together with the assignment of demand, transit frequencies are optimized and the related fleet-size is calculated. Having an optimal setting of solution, the system is continued by reconnecting the routes to find some other better solutions in the periphery of the optimal setting. A set of mathematical programming modules is developed. Real data from Sioux Falls city network is used to evaluate the performance of the model and compare with other heuristic methods.     

Hybrid model of taxonomy and genetic algorithms for finding shortest path in transportation systems

This is research is aimed at elaborating a new methodology of shortest path finding by utilizing the methods of taxonomy and genetic algorithms. Combination of the two is developed and called Genetic Taxonomy Evaluator (GTE) which is expected to be an alternative tool to solve shortest path finding problems within the transportation networks While keeping the properties of transportation networks Taxonomy Reconstructor (TR) transforms the network representation into taxonomic structure, which is hierarchically shaped, based on problem to be solved. In the process TR also creates classification of nodes in the network. This classification provides facilities to isolate the problem to the core, and the criteria that can be inserted in the Genetic Algorithm (GA). A package program for GTE is then developed in C-Language and performance of model is analyzed upon a medium scale of Sioux-Falls City Network. In conclusion, it is found that to achieve fairly quick convergence of GTE computation several optimal parameters of GA should be determined prior to searching for the shortest paths. And since GTE has only been applied to limited case, it is suggested that the findings could be a threshold for further researches.