Two local search approaches for solving real-life car sequencing problems

The NP-hard problem of car sequencing appears as the heart of the logistic process of many car manufacturers. The subject of the ROADEF’2005 challenge addressed a car sequencing problem proposed by the car manufacturer RENAULT, more complex than the academic problem generally addressed in the literature. This paper describes two local search approaches for this problem. In the first part, a new approach by very large-scale neighborhood search is presented. This approach, designed during the qualification stage preceding the final, is based on an original integer linear programming formulation. The second part is dedicated to the approach which enabled us to win the ROADEF’2005 challenge. Inspired by the latest works on the subject, this one is based on very fast explorations of small neighborhoods. Our contribution here is mainly algorithmic, in particular by showing how much exploiting invariants speeds up the neighborhood evaluation and contributes to the diversification of the search. Finally, the two approaches are compared and discussed through an extensive computational study on RENAULT’s benchmarks. The main conclusion drawn at this point is that sophisticated metaheuristics are useless to solve car sequencing problems. More generally, our victory on ROADEF’2005 challenge demonstrates that algorithmic aspects, sometimes neglected, remain the key ingredients for designing and engineering high-performance local search heuristics.     

A hybrid two-stage transportation and batch scheduling problem

We study the coordinated scheduling problem of hybrid batch production on a single batching machine and two-stage transportation connecting the production, where there is a crane available in the first-stage transportation that transports jobs from the warehouse to the machine and there is a vehicle available in the second-stage transportation to deliver jobs from the machine to the customer. As the job to be carried out is big and heavy in the steel industry, it is reasonable assumed that both the crane and the vehicle have unit capacity. The batching machine processes a batch of jobs simultaneously. Each batch occur a setup cost. The objective is to minimize the sum of the makespan and the total setup cost. We prove that this problem is strongly NP-hard. A polynomial time algorithm is proposed for a case where the job transportation times are identical on the crane or the vehicle. An efficient heuristic algorithm for the general problem is constructed and its tight worst-case bound is analyzed. In order to further verify the performance of the proposed heuristics, we develop a lower bound on the optimal objective function. Computational experiments show that the heuristic algorithm performs well on randomly generated problem instances.     

Flow-shop scheduling with setup and assembly operations

A scheduling model for a production system including machining, setup and assembly operations is considered. Production of a number of single-item products is ordered. Each product is made by assembling a set of several different parts. First, the parts are manufactured in a flow-shop consisting of multiple machines. Then, they are assembled into products on a single assembly stage. Setup operation and setup time are needed when a machine starts processing the parts or it changes items. The operations are partitioned into several blocks. Each block consists of the machining operations, the setup operations, and the assembly operation(s) for one or several products. The parts of the same item in a block are processed successively. The objective function is the mean completion time for all products. We consider a problem to partition the operations into blocks and sequence the parts in each block so as to minimize the objective function. Solution procedures using pseudo-dynamic programming and a branch-and-bound method are proposed. Computational experiments are carried out to evaluate the performance of the solution procedures. It has been found that a good near-optimal schedule is obtained efficiently by the proposed solution procedures.     

Minimizing the cycle time of multiple-product processing networks with a fixed operation sequence,setups, and time-window constraints

We solve a special case of the single-robot cyclic scheduling problem with a fixed robot operation sequence and time window constraints on processing times. It generalizes the known single-part fixed-sequence problems into the one to cover a processing network with multiple part types and setup time requirements between the processing steps for different parts at the shared stations. The objective is to minimize the cycle time. We prove that this problem is equivalent to the parametric critical path problem, and propose a strongly polynomial time solution algorithm which uses a new labeling procedure to identify all feasible parameter values. The proposed algorithm is based on an extension to the known Bellman–Ford algorithm.     

Project Scheduling with Multiple Modes: A Genetic Algorithm

In this paper we consider the resource-constrained project scheduling problem with multiple execution modes for each activity and makespan minimization as objective. We present a new genetic algorithm approach to solve this problem. The genetic encoding is based on a precedence feasible list of activities and a mode assignment. After defining the related crossover, mutation, and selection operators, we describe a local search extension which is employed to improve the schedules found by the basic genetic algorithm. Finally, we present the results of our thorough computational study. We determine the best among several different variants of our genetic algorithm and compare it to four other heuristics that have recently been proposed in the literature. The results that have been obtained using a standard set of instances show that the new genetic algorithm outperforms the other heuristic procedures with regard to a lower average deviation from the optimal makespan.     

A coupling cutting stock-lot sizing problem in the paper industry

An important production programming problem arises in paper industries coupling multiple machine scheduling with cutting stocks. Concerning machine scheduling: how can the production of the quantity of large rolls of paper of different types be determined. These rolls are cut to meet demand of items. Scheduling that minimizes setups and production costs may produce rolls which may increase waste in the cutting process. On the other hand, the best number of rolls in the point of view of minimizing waste may lead to high setup costs. In this paper, coupled modeling and heuristic methods are proposed. Computational experiments are presented.     

Scheduling orders on either dedicated or flexible machines in parallel to minimize total weighted completion time

We are interested in the problem of scheduling orders for different product types in a facility with a number of machines in parallel. Each order asks for certain amounts of various different product types which can be produced concurrently. Each product type can be produced on a subset of the machines. Two extreme cases of machine environments are of interest. In the first case, each product type can be produced on one and only one machine which is dedicated to that product type. In the second case, all machines are identical and flexible; each product type can be produced by any one of the machines. Moreover, when a machine in this case switches over from one product type to another, no setup is required. Each order has a release date and a weight. Preemptions are not allowed. The objective is minimizing the total weighted completion time of the orders. Even when all orders are available at time 0, both types of machine environments have been shown to be NP-hard for any fixed number (≥2) of machines. This paper focuses on the design and analysis of approximation algorithms for these two machine environments. We also present empirical comparisons of the various algorithms. The conclusions from the empirical analyses provide insights into the trade-offs with regard to solution quality, speed, and memory space. Electronic Supplementary Material The online version of this article () contains supplementary material, which is available to authorized users. This research is supported by the National Science Foundation through grants DMI-0300156 and DMI-0245603.     

Scheduling networks of queues: Heavy traffic analysis of a simple open network

We consider a queueing network with two single-server stations and two types of customers. Customers of type A require service only at station 1 and customers of type B require service first at station 1 and then at station 2. Each server has a different general service time distribution, and each customer type has a different general interarrival time distribution. The problem is to find a dynamic sequencing policy at station 1 that minimizes the long-run average expected number of customers in the system.The scheduling problem is approximated by a dynamic control problem involving Brownian motion. A reformulation of this control problem is solved, and the solution is interpreted in terms of the queueing system in order to obtain an effective sequencing policy. Also, a pathwise lower bound (for any sequencing policy) is obtained for the total number of customers in the network. We show via simulation that the relative difference between the performance of the proposed policy and the pathwise lower bound becomes small as the load on the network is increased toward the heavy traffic limit.     

A discrete artificial bee colony algorithm for the multi-objective flexible job-shop scheduling problem with maintenance activities

This paper presents a novel discrete artificial bee colony (DABC) algorithm for solving the multi-objective flexible job shop scheduling problem with maintenance activities. Performance criteria considered are the maximum completion time so called makespan, the total workload of machines and the workload of the critical machine. Unlike the original ABC algorithm, the proposed DABC algorithm presents a unique solution representation where a food source is represented by two discrete vectors and tabu search (TS) is applied to each food source to generate neighboring food sources for the employed bees, onlooker bees, and scout bees. An efficient initialization scheme is introduced to construct the initial population with a certain level of quality and diversity. A self-adaptive strategy is adopted to enable the DABC algorithm with learning ability for producing neighboring solutions in different promising regions whereas an external Pareto archive set is designed to record the non-dominated solutions found so far. Furthermore, a novel decoding method is also presented to tackle maintenance activities in schedules generated. The proposed DABC algorithm is tested on a set of the well-known benchmark instances from the existing literature. Through a detailed analysis of experimental results, the highly effective and efficient performance of the proposed DABC algorithm is shown against the best performing algorithms from the literature.     

On the complexity of project scheduling to minimize exposed time

We consider project scheduling where the project manager’s objective is to minimize the time from when an adversary discovers the project until the completion of the project. We analyze the complexity of the problem identifying both polynomially solvable and NP-hard versions of the problem. The complexity of the problem is seen to be dependent on the nature of renewable resource constraints, precedence constraints, and the ability to crash activities in the project.