Distributed algorithm for empty vehicles management in personal rapid transit (PRT) network

📝 Abstract

In this paper, an original heuristic algorithm of empty vehicles management in personal rapid transit network is presented. The algorithm is used for the delivery of empty vehicles for waiting passengers, for balancing the distribution of empty vehicles within the network, and for providing an empty space for vehicles approaching a station. Each of these tasks involves a decision on the trip that has to be done by a selected empty vehicle from its actual location to some determined destination. The decisions are based on a multi-parameter function involving a set of factors and thresholds. An important feature of the algorithm is that it does not use any central database of passenger input (demand) and locations of free vehicles. Instead, it is based on the local exchange of data between stations: on their states and on the vehicles they expect. Therefore, it seems well-tailored for a distributed implementation. The algorithm is uniform, meaning that the same basic procedure is used for multiple tasks using a task-specific set of parameters.

💡 Analysis

In this paper, an original heuristic algorithm of empty vehicles management in personal rapid transit network is presented. The algorithm is used for the delivery of empty vehicles for waiting passengers, for balancing the distribution of empty vehicles within the network, and for providing an empty space for vehicles approaching a station. Each of these tasks involves a decision on the trip that has to be done by a selected empty vehicle from its actual location to some determined destination. The decisions are based on a multi-parameter function involving a set of factors and thresholds. An important feature of the algorithm is that it does not use any central database of passenger input (demand) and locations of free vehicles. Instead, it is based on the local exchange of data between stations: on their states and on the vehicles they expect. Therefore, it seems well-tailored for a distributed implementation. The algorithm is uniform, meaning that the same basic procedure is used for multiple tasks using a task-specific set of parameters.

📄 Content

JOURNAL OF ADVANCED TRANSPORTATION J. Adv. Transp. (2016) Published online in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/atr.1365 http://onlinelibrary.wiley.com/enhanced/doi/10.1002/atr.1365

Distributed Algorithm for Empty Vehicles Management
in Personal Rapid Transit (PRT) network

Wiktor B. Daszczuk, Ph.D. - Institute of Computer Science, Warsaw University of Technology, Nowowiejska str. 15/19, 00-665 Warszawa, Poland, Tel. (+48)22-234-7812, Fax (+48)22-234-6091, Email: wbd@ii.pw.edu.pl,
Jerzy Mieścicki, Ph.D - Institute of Computer Science, Warsaw University of Technology, Nowowiejska str. 15/19, 00-665 Warszawa, Poland, Tel. (+48)22-234-7812, Fax (+48)22-234-6091, Email: jms@ii.pw.edu.pl,
Waldemar Grabski, M.Sc - Institute of Computer Science, Warsaw University of Technology, Nowowiejska str. 15/19, 00-665 Warszawa, Poland, Tel. (+48)22-234-7812, Fax (+48)22-234-6091, Email: wgr@ii.pw.edu.pl Abstract: In this paper, an original heuristic algorithm of empty vehicles management in Personal Rapid Transit (PRT) network is presented. The algorithm is used for the delivery of empty vehicles for waiting passengers; for balancing the distribution of empty vehicles within the network; for providing an empty space for vehicles approaching a station, etc. Each of these tasks involves a decision on the trip which has to be done by a selected empty vehicle from its actual location to some determined destination. The decisions are based on a multi-parameter function, involving a set of factors and thresholds.
An important feature of the algorithm is that it does not use any central database of passenger input (demand) and locations of free vehicles. Instead, it is based on the local exchange of data between stations: on their states and on the vehicles they expect. Therefore, it seems well tailored for a distributed implementation.
The algorithm is uniform, meaning that the same basic procedure is used for multiple tasks using task-specific set of parameters.
Keywords: Personal Rapid Transit; Empty Vehicles Management; Transport Simulation; Transportation Management

Introduction Personal Rapid Transit (PRT) [1,2,3,4] is an urban transit system organized as a network, covering, for instance, a part of a city or a multi-terminal airport, a large exhibition area etc. The driverless (i.e. system-controlled) vehicles move along one-way tracks which are separated from a conventional traffic, e.g., through elevation of the tracks above the ground level. Typically, a vehicle can carry a group of 1 – 6 passengers. The term ‘personal’ means that a passenger, or a group of passengers, chooses the time as well as the destination of a trip freely. The system determines the best route for the trip, which is not necessarily the shortest one, and controls the vehicle movement during the voyage (acceleration/deceleration, preserving the separation between vehicles, passing intersections, avoiding traffic jams, etc.) . In addition to the control of these so-called full trips, the system manages the set of empty vehicles movement as well, which is analyzed in more detail below. We can divide control algorithms of the PRT network into two levels:  Coordination algorithms, used for the control of vehicles’ movement following one another down the track, for the coordination at join-type intersections and for the control of the vehicle behavior inside stations and capacitors, as described in [5]. These coordination algorithms are not a subject of this paper.  Management algorithms, including empty vehicle management and dynamic routing algorithms. In the present paper the former feature is addressed. Many algorithms for empty vehicle management known from the literature are focused mainly on the optimal reallocation of empty vehicles, since this may reduce the passenger waiting time significantly (at the expense of increased empty vehicles movement). Reallocation algorithms are usually based on past demand estimates and future forecast [6-14]. All these approaches use a form of central repository in which historical demand and actual empty vehicles supply are stored. The demand forecasts are based on statistics from previous corresponding periods corrected for weather and special events. Due to the central data base, the algorithms are centralized. They mainly refer to empty vehicle reallocation (other cases are addressed separately) and hardly can be implemented in a distributed way.
Some papers [8,14] take into account other features, like the distance between stations or the time in which a vehicle may be delivered (in fact, this limits the distance as well). Yet, none gives a set of factors and thresholds to tune an algorithm to a given network with a given number of vehicles and a given demand. The work presented in [12,13] is extended in [15], where the traffic needed to move full and empty vehicles to their destinations is limited by the

This content is AI-processed based on ArXiv data.