A Decentralised Approach for the Transportation On Demand Problem

Abstract

Public transport systems are generally organized in a static, a priori way. In such systems, the demand must be adapted to the offer. In this paper, we propose a model based on self-organization in order to dispatch a fleet of vehicles in a purely dynamic Transportation On Demand system (TOD). Our proposal consists in a decentralized approach and a multi-agent system (MAS) to model the environment. This will tackle the problem of vehicles over-concentration or the lack of service in certain areas of the city. We demonstrate that our model addresses these problems by providing vehicle agents, for a given request, to make the final decision thanks to a negotiation process and to calculate overcosts according to an original insertion heuristic.

A survey on demand-responsive public bus systems

“…The authors conclude that the field tests show that the heuristic can handle large scale problems and customers perceived the system as practical and efficient. The work of Bertelle et al (2009) stands in contrast with the previous because this work investigates a theoretical decentralized DON-PBS. The system emerges from the present fleet of vehicles.…”

“…Many-to-many fully flexible DON-PBS are also frequently optimized from the operator's perspective, mostly to make sure the system is profitable. A stop-based approach with zero lead time is adopted by Tsubouchi et al (2010) and Bertelle et al (2009). The first build a heuristic based on efficient vehicle choosing and online insertion and time adjustment.…”

Smart Method for Self-Organization in Last-Mile Parcel Delivery

“…However, real-life large-scale dynamic problems with many transportation requests are almost impossible to solve quickly. Bertelle et al ( 28 ) have introduced a decentralized approach resembling the ECNP to solve the dynamic transportation on demand problem using three agent types: clients, vehicles, and a central interface. On receiving a transportation request, vehicles use an insertion heuristic to calculate the additional effort of accepting that request.…”

A Multi-agent Intelligent System for on Demand Transport Problem Solving

“…Finally, in (Bertelle, Nabaa, Olivier and Tranouez 2009) develop a decentralized approach based on the optimization and negotiation between vehicles to resolve the ODT problem. Their goal is to face the lack of service in certain zones or the over-concentration of vehicles in certain other zones in a changing environment.…”