Tiente Hsu

Cyclic scheduling for F.M.S : Modelling and evolutionary solving approach

This paper concerns the domain of flexible manufacturing systems (FMS) and focuses on the scheduling problems encountered in these systems. We have chosen the cyclic behaviour to study this problem, to reduce its complexity. This cyclic scheduling problem, whose complexity is NP-hard in the general case, aims to minimise the work in process (WIP) to satisfy economic constraints. We first recall and discuss the best known cyclic scheduling heuristics. Then, we present a two-step resolution approach. In the first step, a performance analysis is carried out; it is based on the Petri net modelling of the production process. This analysis resolves some indeterminism due to the systems flexibility and allows a lower bound of the WIP to be obtained. In the second step, after a formal model of the scheduling problem has been given, we describe a genetic algorithm approach to find a schedule which can reach the optimal production speed while minimizing the WIP. Finally, our genetic approach is validated and compared with known heuristics on a set of test problems.

A hybrid GA approach for solving the Dynamic Vehicle Routing Problem with Time Windows

The dynamic vehicle routing problem is an extension of conventional routing problems whose particularities are that information can change after initial routes have been constructed or that not all the information is known when the routing process takes place. The main interest of this type of problem is that it corresponds to many real word applications (repair services, courier mail services, taxi cab services). In this paper, we study the particular case of the dynamic vehicle routing problem with time windows (DVRPTW) in which occurrences of new customers appear over time. We propose an original resolution approach based on a genetic algorithm adapted to this dynamic optimisation context. Taguchis tables have been used in order to adjust the parameters of our genetic algorithm. Experimental results based on the modified Solomon benchmarks show the efficacy of our approach as compared to other meta-heuristic approaches

Vehicle routing problem with time windows and fuzzy demands: an approach based on the possibility theory

In this paper, the vehicle routing problem with time windows and fuzzy demands (VRPTWFD) is considered and a fuzzy recourse model based on the possibility theory is proposed. A stochastic simulation and a genetic algorithm (GA) are integrated to design a hybrid intelligent algorithm to solve the fuzzy version of a stochastic recourse model. Moreover, an adaptation of the Solomons benchmark is proposed in order to assess the quality of the proposed approach.

A Multi-agent Based Self-adaptive Genetic Algorithm for the Long-term Car Pooling Problem

Rising vehicles number and increased use of private cars have caused significant traffic congestion, noise and energy waste. Public transport cannot always be set up in the non-urban areas. Car pooling, which is based on the idea that sets of car owners having the same travel destination share their vehicles has emerged to be a viable possibility to reduce private car usage around the world. In this paper, we present a multi-agent based self-adaptive genetic algorithm to solve long-term car pooling problem. The system is a combination of multi-agent system and genetic paradigm, and guided by a hyper-heuristic dynamically adapted by a collective learning process. The aim of our research is to solve the long-term car pooling problem efficiently with limited exploration of the search space. The proposed algorithm is tested using large scale instance data sets. The computational results show that the proposed method is competitive with other known approaches for solving long-term car pooling problem.

Genetic algorithm for the vehicle routing problem with time windows and fuzzy demand

This paper considers a VRP with soft time windows and fuzzy demand (VRPTWFD). The objective is to minimize both the total distance covered by all vehicles as well as the sum of lateness at the customerpsilas due to the violation of time windows. This VRPTWFD is formulated as a two stages recourse model in the context of stochastic programming. The goal is then to minimize the expected cost, which includes the initial cost of the solution found in first stage and the additional cost due to the route failure in second stage. The theory of possibility is applied in the capacity constraint. In addition, a route failure estimation method is proposed to evaluate the additional cost as well as the expected cost. A genetic algorithm, in which a simulation phase based on sampling scenarios to evaluate the fitness of chromosome, is specifically designed to solve the two stages recourse model for the VRPTWFD. Finally an experimental evaluation of this developed algorithm is validated on a few VRPTWFD modified from the Solomon benchmarks.

Dynamic Vehicle Routing Problems under Flexible Time Windows and Fuzzy Travel Times

The current transport systems have to provide a given level of service to the final customer. Dynamic piloting models, able to react quickly to the events must be designed to guarantee this level of service. In this article, we are interested in the dynamic management of a vehicles fleet, dedicated to the pick-up and/or delivery of parcels in a customers set. In order to have a more realistic model, travel times and time windows are subject to variations. We present two indicators qualifying a vehicle routing service level in such situation. They allow the decision maker to realize a compromise between service quality and risk taking. These indicators then integrated in our platform, to solve the dynamic problem under uncertainties

Embedding trees in massively parallel computers

Abstract Trees are a useful class of computational structures. In parallel processing, mapping problems arise when the tree structure differs from the processor interconnection of the parallel computer. The MasPars MP-2 is a massively parallel computer where processors are interconnected via the X-Net neighborhood two-dimensional mesh and the global multistage crossbar router network. We present novel mapping schemes for trees on the MasPars MP-2 two-dimensional mesh and on the two-dimensional mesh together with the MasPars MP-2 multistage crossbar network. Appropriate algorithms of the mapping scheme on the two-dimensional mesh (or grid) are presented and are shown to be superior over known mappings on square arrays (or grids).

Embedding tree structures in massively parallel computers

Trees are an important class of computational structures due to their favoumble property of logarithmic path from the root to any leaf element. Since simple and regular interconnections among nodes are needed, they are easily manipulated. Mapping problems arise when the tree structure differs from the processor interconnection of the pamllel computer. The MasPar’s MP-2 is a massively parallel computer where processors are interconnected via the X-Net neighborhood two-dimensional mesh and the globul multistage crossbar router network. We present novel mapping schemes for trees on the MasPar’s MP-2 twodimensional mesh and on the two-dimensional mesh together with the MasPar’s MP-2 multistage crossbar network. Appropriate algorithms of the mapping scheme on the two-dimensional mesh (or grid) are presented and are shown to be superior over known mappings on square arrays (or grids). keywords: massively parallel systems, interconnection networks, mapping scheme, tree structures, distributed configuration algorithm, evaluation.

A multi-destination daily carpooling problem and an ant colony based resolution method

The rising car usage deriving from growth in jobs and residential population causes air pollution, energy waste and consumption of people’s time. Public transport cannot be the only answer to this increasing transport demand. Carpooling, which is based on the idea that sets of car owners pick up colleagues while driving to or from the workplace, has emerged to be a viable possibility for reducing private car usage in congested areas. Its actual practice requires a suitable information system support and, the most important, the capability of effectively solving the underlying combinatorial optimization problem. This paper describes an ant colony algorithm based hybrid approach (HAC) for solving the multi-destination carpooling problem. Experiments have been performed to confirm the efficiency and the effectiveness of the approach.

Iterated Local Search and Column Generation to solve Arc-Routing as a permutation set-covering problem

We propose a method that combines Column Generation (CG) and Iterated Local Search (ILS) to solve the Capacitated Arc-Routing Problem (CARP). One of the goals is to integrate into the ILS (some of) the duality information that underpins the CG paradigm. The CARP is expressed in a space of permutations and sub-permutations (sequences, routes) of the set of required edges. For this, the ILS uses an exact decoder that maps any permutation s to a list of sequences (routes) of minimum cost servicing all edges in the order s(1), s(2), s(3), etc. This permutation space is explored both by an ILS process and a CG process that run in parallel and that communicate by exchanging sequences. The first use of the CG paradigm in ILS is the following: all sequences discovered by CG are sent to the ILS process that can inject them into the current ILS solution. The second application of CG in ILS is a “CG improver” operator that acts on the current ILS solution, so as to (try to) improve it by running several CG iterations. The first half of the paper describes the method in a general framework based on sequences, permutations and set covering. The second part is devoted to more specialized Arc-Routing techniques. For instance, the CG convergence could be accelerated by factors of tens or even hundreds by exploiting two ideas in the Dynamic Programming (DP) pricing: (i) avoid as much as possible to traverse edges without service, and (ii) record only non-dominated DP states using a fast-access data structure mixing an array and a red-black tree. Regarding the ILS, we show that the permutation-level search can be substantially improved if the exact decoder is reinforced by a deterministic post-decoder acting on explicit routes. The general results are competitive (reducing the best-known gap of five instances) and certain ideas could be potentially useful for other set-covering or permutation problems.