Ferdinando Pezzella

A tabu search method guided by shifting bottleneck for the job shop scheduling problem

Abstract A computationally effective heuristic method for solving the minimum makespan problem of job shop scheduling is presented. The proposed local search method is based on a tabu search technique and on the shifting bottleneck procedure used to generate the initial solution and to refine the next-current solutions. Computational experiments on a standard set of problem instances show that, in several cases, our approach, in a reasonable amount of computer time, yields better results than the other heuristic procedures discussed in the literature.

An Improved Genetic Algorithm for the Distributed and Flexible Job-shop Scheduling problem

The Distributed and Flexible Job-shop Scheduling problem (DFJS) considers the scheduling of distributed manufacturing environments, where jobs are processed by a system of several Flexible Manufacturing Units (FMUs). Distributed scheduling problems deal with the assignment of jobs to FMUs and with determining the scheduling of each FMU, in terms of assignment of each job operation to one of the machines able to work it (job-routing flexibility) and sequence of operations on each machine. The objective is to minimize the global makespan over all the FMUs. This paper proposes an Improved Genetic Algorithm to solve the Distributed and Flexible Job-shop Scheduling problem. With respect to the solution representation for non-distributed job-shop scheduling, gene encoding is extended to include information on job-to-FMU assignment, and a greedy decoding procedure exploits flexibility and determines the job routings. Besides traditional crossover and mutation operators, a new local search based operator is used to improve available solutions by refining the most promising individuals of each generation. The proposed approach has been compared with other algorithms for distributed scheduling and evaluated with satisfactory results on a large set of distributed-and-flexible scheduling problems derived from classical job-shop scheduling benchmarks.

A Variable Neighborhood Search Branching for the Electric Vehicle Routing Problem with Time Windows

Abstract E-mobility plays a key role especially in contexts where the transportation activities impact a lot on the total costs. The Electric Vehicles (EVs) are becoming an effective alternative to the internal combustion engines guaranteeing cheaper and eco-sustainable transport solutions. However, the poor battery autonomy is still an Achilles hell since the EVs require many stops for being recharged. We aim to optimally route the EVs for handling a set of customers in time considering the recharging needs during the trips. A Mixed Integer Linear Programming formulation of the problem is proposed assuming that the battery recharging level reached at each station is a decision variable in order to guarantee more flexible routes. The model aims to minimize the total travel, waiting and recharging time plus the number of the employed EVs. Finally, a Variable Neighborhood Search Branching (VNSB) is also designed for solving the problem at hand in reasonable computational times. Numerical results on benchmark instances show the performances of both the mathematical formulation and the VNSB compared to the ones of the model in which the battery level reached at each station is always equal to the maximum capacity.

EVE-OPT: a hybrid algorithm for the capacitated vehicle routing problem

This paper presents EVE-OPT, a Hybrid Algorithm based on Genetic Algorithms and Taboo Search for solving the Capacitated Vehicle Routing Problem. Several hybrid algorithms have been proposed in recent years for solving this problem. Despite good results, they usually make use of highly problem-dependent neighbourhoods and complex genetic operators. This makes their application to real instances difficult, as a number of additional constraints need to be considered. The algorithm described here hybridizes two very simple heuristics and introduces a new genetic operator, the Chain Mutation, as well as a new mutation scheme. We also apply a procedure, the k-chain-moves, able to increase the neighbourhood size, thereby improving the quality of the solution with negligible computational effort. Despite its simplicity, EVE-OPT is able to achieve the same results as very complex state-of-the art algorithms.

An adaptive genetic algorithm for large-size open stack problems

The problem of minimising the maximum number of open stacks arises in many contexts (production planning, cutting environments, very-large-scale-integration circuit design, etc.) and consists of finding a sequence of tasks (products, cutting patterns, circuit gates, etc.) that determines an efficient utilisation of resources (stacks). We propose a genetic approach that combines classical genetic operators (selection, order crossover and pairwise interchange mutation) with an adaptive search strategy, where intensification and diversification phases are obtained by neighbourhood search and by a composite and dynamic fitness function that suitably modifies the search landscape. Computational tests on random and real-world benchmarks show that the proposed approach is competitive with the state of the art for large-size problems, providing better results for some classes of instances.

Heuristic algorithms for the operator-based relocation problem in one-way electric carsharing systems

This paper addresses an Electric Vehicle Relocation Problem (E-VReP), in one-way carsharing systems, based on operators who use folding bicycles to facilitate vehicle relocation. In order to calculate the economic sustainability of this relocation approach, a revenue associated with each relocation request satisfied and a cost associated with each operator used are introduced. The new optimization objective maximizes the total profit. To overcome the drawback of the high CPU time required by the Mixed Integer Linear Programming formulation of the E-VReP, two heuristic algorithms, based on the general properties of the feasible solutions, are designed. Their effectiveness is tested on two sets of realistic instances. In the first, all the requests have the same revenue, while, in the second, the revenue of each request has a variable component related to the users rent-time and a fixed part related to customer satisfaction. Finally, a sensitivity analysis is carried out on both the number of requests and the fixed revenue component. Economic sustainability of E-VReP in one way carsharing systems was addressed.Computational complexity and APX-hardness is addressed.Heuristics were designed to overcome the drawback of the high CPU times of the MILP.Variable revenues associated with requests satisfied were also considered.Sensitivity analysis was carried out on the number of requests and the revenue.

A new Mathematical Programming Model for the Green Vehicle Routing Problem

Abstract A new MILP formulation for the Green Vehicle Routing Problem is introduced where the visits to the Alternative Fuel Stations (AFSs) are only implicitly considered. The number of variables is also reduced by pre-computing for each couple of customers an efficient set of AFSs, only given by those that may be actually used in an optimal solution. Numerical experiments on benchmark instances show that our model outperforms the previous ones proposed in the literature.

An Adaptive Large Neighborhood Search for relocating vehicles in electric carsharing services

Abstract We propose an Adaptive Large Neighborhood Search metaheuristic to solve a vehicle relocation problem arising in the management of electric carsharing systems. The solution approach, aimed to optimize the total profit, is tested on three real-like benchmark sets of instances. It is compared with a Tabu Search, ad hoc designed for this work, with a previous Ruin and Recreate metaheuristic and with the optimal results obtained via Mixed Integer Linear Programming. We also develop bounding procedures to evaluate the solution quality when the optimal solution is not available.

A three-phase matheuristic for the time-effective electric vehicle routing problem with partial recharges

Abstract We propose a three-phase matheuristic, combining an exact method with a Variable Neighborhood Search local Branching (VNSB) to route a fleet of Electric Vehicles (EVs). EVs are allowed stopping at the recharging stations along their routes to (also partially) recharge their batteries. We hierarchically minimize the number of EVs used and the total time spent by the EVs, i.e., travel times, charging times and waiting times (due to the customer time windows). The first two phases are based on Mixed Integer Linear Programs to generate feasible solutions, used in a VNSB algorithm. Numerical results on benchmark instances show that the proposed approach finds good quality solutions in reasonable amount of time.

A two-phase optimization method for a multiobjective vehicle relocation problem in electric carsharing systems

The paper focuses on one-way electric carsharing systems, where the fleet of cars is made up of Electric Vehicles (EVs) and the users can pick-up the EV at a station and return it to a different one. Such systems require efficient vehicle relocation for constantly balancing the availability of EVs among stations. In this work, the EVs are relocated by workers, and the issue of finding a trade-off among the customers’ satisfaction, the workers’ workload balance and the carsharing provider’s objective is addressed. This leads to a three-objective optimization problem for which a two-phase solution approach is proposed. In the first phase, feasible routes and schedules for relocating EVs are generated by different randomized search heuristics; in the second phase, non-dominated solutions are found through epsilon-constraint programming. Computational results are performed on benchmark instances and new large size instances based on the city of Milan.