Energetic reasoning and bin-packing problem, for bounding a parallel machine scheduling problem

This paper deals with an extension of energetic reasoning, using some efficient lower bounds of the bin-packing problem, to get tight lower bounds for the P|r _i , q _i |C _max. The link between P||C _max and bin-packing problem is well-known. Our purpose is to extend the use of efficient lower bounds of the bin-packing problem to P|r _i , q _i |C _max. We focus on some time-intervals, to compute the mandatory parts of activities within this time-interval and then to deduce an associated bin-packing instance. Thus, lower bounds of the bin-packing problem are used to get new satisfiability tests for the parallel machine problem. We also propose to extend the classical time-bound adjustments of release dates and deadlines to efficiently use bin-packing lower bounds. Experimental results that prove the efficiency of our approach on several kind of instances are reported.     

Divide-and-price: A decomposition algorithm for solving large railway crew scheduling problems

The railway crew scheduling problem consists of generating crew duties to operate trains at minimal cost, while meeting all work regulations and operational requirements. Typically, a railway operation uses tens of thousands of train movements (trips) and requires thousands of crew members to be assigned to these trips. Despite the large size of the problem, crew schedules need to be generated in short time, because large parts of the train schedule are not finalized until few days before operation.     

Parallel machine scheduling problems with a single server

In this paper, we consider the problem of scheduling jobs on parallel machines with setup times. The setup has to be performed by a single server. The objective is to minimize the schedule length (makespan), as well as the forced idle time. The makespan problem is known to be NP-hard even for the case of two identical parallel machines. This paper presents a pseudopolynomial algorithm for the case of two machines when all setup times are equal to one. We also show that the more general problem with an arbitrary number of machines is unary NP-hard and analyze some list scheduling heuristics for this problem. The problem of minimizing the forced idle time is known to be unary NP-hard for the case of two machines and arbitrary setup and processing times. We prove unary NP-hardness of this problem even for the case of constant setup times. Moreover, some polynomially solvable cases are given.     

Design of cooperative algorithms for multi-objective optimization: application to the flow-shop scheduling problem

This is a summary of the main results presented in the author’s PhD thesis. This thesis was supervised by El-Ghazali Talbi, and defended on 21 June 2005 at the University of Lille (France). It is written in French and is available at http://www.lifl.fr/~basseur/These.pdf. This work deals with the conception of cooperative methods in order to solve multi-objective combinatorial optimization problems. Many cooperation schemes between exact and/or heuristic methods have been proposed in the literature. We propose a classification of such schemes. We propose a new heuristic called adaptive genetic algorithm (AGA), that is designed for an efficient exploration of the search space. We consider several cooperation schemes between AGA and other methods (exact or heuristic). The performance of these schemes are tested on a bi-objective permutation flow-shop scheduling problem, in order to evaluate the interest of each type of cooperation.     

Domination analysis for minimum multiprocessor scheduling

Let P be a combinatorial optimization problem, and let A be an approximation algorithm for P. The domination ratio domr(A,s) is the maximal real q such that the solution x(I) obtained by A for any instance I of P of size s is not worse than at least the fraction q of the feasible solutions of I. We say that P admits an asymptotic domination ratio one (ADRO) algorithm if there is a polynomial time approximation algorithm A for P such that . Alon, Gutin and Krivelevich [Algorithms with large domination ratio, J. Algorithms 50 (2004) 118-131] proved that the partition problem admits an ADRO algorithm. We extend their result to the minimum multiprocessor scheduling problem.     

Heuristic algorithms for siting alternative-fuel stations using the Flow-Refueling Location Model

This paper presents three heuristic algorithms that solve for the optimal locations for refueling stations for alternative-fuels, such as hydrogen, ethanol, biodiesel, natural gas, or electricity. The Flow-Refueling Location Model (FRLM) locates refueling stations to maximize the flow that can be refueled with a given number of facilities. The FRLM uses path-based demands, and because of the limitations imposed by the driving range of vehicles, longer paths require combinations of more than one station to refuel round-trip travel. A mixed-integer linear programming (MILP) version of the model has been formulated and published and could be used to obtain an optimal solution. However, because of the need for combinations of stations to satisfy demands, a realistic problem with a moderate size network and a reasonable number of candidate sites would be impractical to generate and solve with MILP methods. In this research, heuristic algorithms—specifically the greedy-adding, greedy-adding with substitution and genetic algorithm—are developed and applied to solve the FRLM problem. These algorithms are shown to be effective and efficient in solving complex FRLM problems. For case study purposes, the heuristic algorithms are applied to locate hydrogen-refueling stations in the state of Florida.     

Effective learning hyper-heuristics for the course timetabling problem

Course timetabling is an important and recurring administrative activity in most educational institutions. This article combines a general modeling methodology with effective learning hyper-heuristics to solve this problem. The proposed hyper-heuristics are based on an iterated local search procedure that autonomously combines a set of move operators. Two types of learning for operator selection are contrasted: a static (offline) approach, with a clear distinction between training and execution phases; and a dynamic approach that learns on the fly. The resulting algorithms are tested over the set of real-world instances collected by the first and second International Timetabling competitions. The dynamic scheme statistically outperforms the static counterpart, and produces competitive results when compared to the state-of-the-art, even producing a new best-known solution. Importantly, our study illustrates that algorithms with increased autonomy and generality can outperform human designed problem-specific algorithms.     

Efficient local search algorithms for known and new neighborhoods for the generalized traveling salesman problem

The generalized traveling salesman problem (GTSP) is a well-known combinatorial optimization problem with a host of applications. It is an extension of the Traveling Salesman Problem (TSP) where the set of cities is partitioned into so-called clusters, and the salesman has to visit every cluster exactly once.     

Using clustering analysis in a capacitated location-routing problem

The location routing problem (LRP) appears as a combination of two difficult problems: the facility location problem (FLP) and the vehicle routing problem (VRP). In this work, we consider a discrete LRP with two levels: a set of potential capacitated distribution centres (DC) and a set of ordered customers. In our problem we intend to determine the set of installed DCs as well as the distribution routes (starting and ending at the DC). The problem is also constrained with capacities on the vehicles. Moreover, there is a homogeneous fleet of vehicles, carrying a single product and each customer is visited just once. As an objective we intend to minimize the routing and location costs.     

Minimizing the number of machines for scheduling jobs with equal processing times

In this paper, we consider a parallel machine environment when all jobs have the same processing time and arbitrary release dates and deadlines of the jobs are given. We suppose that the available number of machines, which can be used simultaneously, may vary over time. The aim is to construct a feasible schedule in such a way that the maximal number of simultaneously used machines is minimal. We give a polynomial algorithm for this problem.     

A caving degree approach for the single container loading problem

Inspired by an old adage “Gold corner, silver side and strawy void”, and improved by a new observation “Maximum value in diamond cave”, a new heuristic approach is proposed for solving the three-dimensional single container loading problem. Differing from several previous approaches, its key issue is to pack the outside item into a corner or even a cave in the container such that the item is as compactly and closely as possible with other packed items. Experiments are on two groups of public and difficult benchmarks. For the 47 without-orientation-constraint instances from the OR-Library, experiments indicate an average packing utilization of 94.9%, which improves current best result reported in the literature by 3.9%. For the 800 strongly heterogeneous instances among 1500 representative benchmarks proposed by Bischoff et al., (100 instances in a set), experiments show an average packing utilization of 87.97%, which improves current best record reported in the literature by 0.28%. Besides, new best records are achieved on the latter five sets among the eight sets of strongly heterogeneous benchmarks.     

Genetic algorithm based approach for the integrated airline crew-pairing and rostering problem

Airline crew scheduling problem is a complex and difficult problem faced by all airline companies.To tackle this problem, it was often decomposed into two subproblems solved successively. First, the airline crew-pairing problem, which consists on finding a set of trips – called pairings – i.e. sequences of flights, starting and ending at a crew base, that cover all the flights planned for a given period of time. Secondly, the airline crew rostering problem, which consists on assigning the pairings found by solving the first subproblem, to the named airline crew members. For both problems, several rules and regulations must be respected and costs minimized.It is sure that this decomposition provides a convenient tool to handle the numerous and complex restrictions, but it lacks, however, of a global treatment of the problem. For this purpose, in this study we took the challenge of proposing a new way to solve both subproblems simultaneously. The proposed approach is based on a hybrid genetic algorithm. In fact, three heuristics are developed here to tackle the restriction rules within the GA’s process.     

A hybrid adaptively genetic algorithm for task scheduling problem in the phased array radar

A phased array radar (PAR) is used to detect new targets and update the information of those detected targets. Generally, a large number of tasks need to be performed by a single PAR in a finite time horizon. In order to utilize the limited time and the energy resources, it is necessary to provide an efficient task scheduling algorithm. However, the existing radar task scheduling algorithms can't be utilized to release the full potential of the PAR, because of those disadvantages such as full PAR task structure ignored, only good performance in one aspect considered and just heuristic or the meta-heuristic method utilized. Aiming at above issues, an optimization model for the PAR task scheduling and a hybrid adaptively genetic (HAGA) algorithm are proposed. The model considers the full PAR task structure and integrates multiple principles of task scheduling, so that multi-aspect performance can be guaranteed. The HAGA incorporates the improved GA to explore better solutions while using the heuristic task interleaving algorithm to utilize wait intervals to interleave subtasks and calculate fitness values of individuals in efficient manners. Furthermore, the efficiency and the effectiveness of the HAGA are both improved by adopting chaotic sequences for the population initialization, the elite reservation and the mixed ranking selection, as well as designing the adaptive crossover and the adaptive mutation operators. The simulation results demonstrate that the HAGA possesses merits of global exploration, faster convergence, and robustness compared with three state-of-art algorithms—adaptive GA, hybrid GA and highest priority and earliest deadline first heuristic (HPEDF) algorithm.     

A dynamic adaptive local search algorithm for the circular packing problem

This paper studies the circular packing problem (CPP) which consists of packing n non-identical circles C_i of known radius r_i, i ∈ N = {1, … , n}, into the smallest containing circle C. The objective is to determine the coordinates (x_i, y_i) of the center of C_i, i ∈ N, as well as the radius r and center (x, y) of C. This problem, which is a variant of the two-dimensional open dimension problem, is solved using a two-step, dynamic, adaptive, local search algorithm. At each iteration, the algorithm identifies the set of potential “best local positions” of a circle C_i, i ∈ N, given the positions of the previously packed circles, and determines for each of these positions the coordinates and radius of the smallest containing circle. The “best local position” minimizes the radius of the current containing circle. That is, every time an additional circle is packed, both the center and the radius of the containing circle are dynamically updated, and the smallest containing circle is known. The experimental results reflect the good performance of the algorithm.     

Approximation of the supply scheduling problem

The supply scheduling problem consists in finding a minimum cost delivery plan from a set of providers to a manufacturing unit, subject to given bounds on the shipment sizes and subject to the demand at the manufacturing unit. We provide a fully polynomial time approximation scheme for this problem.     

Optimal receiver scheduling algorithms for a multicast problem

This paper studies a multicast problem arising in wavelength division multiplexing single-hop lightwave networks, as well as in Video-on-Demand systems. In this problem, the same message of duration Δ has to be transmitted to a set of n receivers which are not all available simultaneously. The receivers can be partitioned into subsets, each served by a different transmission, with the objective of minimizing their overall waiting cost. When there is a single data channel available for transmission, a dynamic programming algorithm is devised which finds an optimal solution in O(nlogn+min{n²,nΔ²}) time, improving over a previously known O(n³) time algorithm. When multiple data channels are available for transmission, an optimal O(n) time algorithm is proposed which finds an optimal solution if the message has constant transmission duration, whereas an NP-completeness proof is given if the message has arbitrary transmission duration.     

A hierarchical scheduling problem with a well-solvable second stage

In the hierarchical scheduling model to be considered, the decision at the aggregate level to acquire a number of identical machines has to be based on probabilistic information about the jobs that have to be scheduled on these machines at the detailed level. The objective is to minimize the sum of the acquisition costs and the expected average completion time of the jobs. In contrast to previous models of this type, the second part of this objective function corresponds to a well-solvable scheduling problem that can be solved to optimality by a simple priority rule. A heuristic method to solve the entire problem is described, for which strong asymptotic optimality results can be established.     

The course scheduling problem at Lufthansa Technical Training

Lufthansa Technical Training GmbH (LTT) performs training courses for Lufthansa Technik AG as well as for several other international airlines. Courses of about 670 different types are offered of which several hundred take place each year. The course scheduling problem faced by LTT is to construct a yearly schedule which maximizes the profit margin incurred while meeting a variety of complex precedence, temporal, and resource-related constraints. A “good” operational schedule should also meet a number of additional subordinate objectives. We formalize the problem and develop a heuristic scheme along with several priority rules, as well as a local search algorithm to determine well-suited weights for weighted composite rules. The operational planning situation of 1996 served as our major test instance; additional test instances were constructed by modifying this data. Several computational experiments were carried out to evaluate the performance of the algorithms. It turned out that the best so-found schedule is substantially better in terms of the profit margin incurred than the solution manually constructed by LTT.     

General solutions to the single vehicle routing problem with pickups and deliveries

The single vehicle routing problem with pickups and deliveries (SVRPPD) is defined on a graph in which pickup and delivery demands are associated with the customer vertices. The problem consists of designing a least cost route for a vehicle of capacity Q. Each customer is allowed to be visited once for a combined pickup and delivery, or twice if these two operations are performed separately. This article proposes a mixed integer linear programming model for the SVRPPD. It introduces the concept of general solution which encompasses known solution shapes such as Hamiltonian, double-path and lasso. Classical construction and improvement heuristics, as well as a tabu search heuristic, are developed and tested over several instances. Computational results show that the best solutions generated by the heuristics are frequently non-Hamiltonian and may contain up to two customers visited twice.     

Approximate solution of a resource-constrained scheduling problem

This paper is devoted to the approximate solution of a strongly NP-hard resource-constrained scheduling problem which arises in relation to the operability of certain high availability real time distributed systems. We present an algorithm based on the simulated annealing metaheuristic and, building on previous research on exact solution methods, extensive computational results demonstrating its practical ability to produce acceptable solutions, in a precisely defined sense. Additionally, our experiments are in remarkable agreement with certain theoretical properties of our simulated annealing scheme. The paper concludes with a short discussion on further research. This research was supported in part by Association Nationale de la Recherche Technique grant CIFRE-121/2004.