Showing papers by "Mario Vanhoucke published in 2005"

An electromagnetic meta-heuristic for the nurse scheduling problem

Abstract: In this paper, we present a novel meta-heuristic technique for the nurse scheduling problem (NSP). This well-known scheduling problem assigns nurses to shifts per day maximizing the overall quality of the roster while taking various constraints into account. The problem is known to be NP-hard. Due to its complexity and relevance, many algorithms have been developed to solve practical and often case-specific models of the NSP. The huge variety of constraints and the several objective function possibilities have led to exact and meta-heuristic procedures in various guises, and hence comparison and state-of-the-art reporting of standard results seem to be a utopian idea. We present a meta-heuristic procedure for the NSP based on the framework proposed by Birbil and Fang (J. Glob. Opt. 25, 263---282, 2003). The Electromagnetic (EM) approach is based on the theory of physics, and simulates attraction and repulsion of sample points in order to move towards a promising solution. Moreover, we present computational experiments on a standard benchmark dataset, and solve problem instances under different assumptions. We show that the proposed procedure performs consistently well under many different circumstances, and hence, can be considered as robust against case-specific constraints.

The project scheduling game (PSG): simulating time/cost trade-offs in projects

Abstract: The Project Scheduling Game is an IT-supported simulation game that illustrates the complexity of scheduling a real-life project. The project is based on a sequence of activities for a large real-l...

A bi-population based genetic algorithm for the resource-constrained project scheduling problem

Abstract: The resource-constrained project scheduling problem (RCP- SP) is one of the most challenging problems in project scheduling. During the last couple of years many heuristic procedures have been developed for this problem, but still these procedures often fail in finding near-optimal solutions for more challenging problem instances. In this paper, we present a new genetic algorithm (GA) that, in contrast of a conventional GA, makes use of two separate populations. This bi-population genetic algorithm (BPGA) operates on both a population of left-justified schedules and a population of right-justified schedules in order to fully exploit the features of the iterative forward/backward scheduling technique. Comparative computational results reveal that this procedure can be considered as today's best performing RCPSP heuristic.

The discrete time/cost trade-off problem under various assumptions exact and heuristic procedures

Abstract: Time/cost trade-offs in project networks have been the subject of extensive research since the development of the critical path method (CPM) in the late 50s. Time/cost behaviour in a project activity basically describes the trade-off between the duration of the activity and its amount of non-renewable resources (e.g. money) committed to it. In the discrete version of the problem (the discrete time/cost trade-off problem), it is generally accepted that the trade-off follows a discrete non-increasing pattern, i.e. expediting an activity is possible by allocating more resources (i.e. at a larger cost) to it. However, due to its complexity (the problem is known to be NP hard (see De et al. (1997)), the problem has been solved for relatively small instances. In this paper, we elaborate on three extensions of the well-known discrete time/cost trade-off problem in order to cope with more realistic settings: time/switch constraints, work continuity constraints and net present value maximization. We give an extensive literature overview of existing procedures for these problem types, and present an exact solution approach for the work continuity version, which is not being investigated yet. Moreover, we discuss a new meta-heuristic approach in order to provide near-optimal heuristic solutions for the different problems. We present computational results for the problems under study by comparing the results for both exact and heuristic procedures. We demonstrate that the heuristic algorithms produce consistently good results for two versions of the discrete time/cost trade-off problem.

A Decomposition-Based Heuristic For The Resource-Constrained Project Scheduling Problem

Abstract: In the last few decades the resource-constrained project scheduling problem has become a popular problem type in operations research. However, due to its strongly NP-hard status, the effectiveness of exact optimisation procedures is restricted to relatively small instances. In this paper we present a new genetic algorithm (GA) for this problem, able to provide near-optimal heuristic solutions. This GA procedure has been extended by a so-called decomposition-based heuristic (DBH) which iteratively solves subparts of the project. We present computational experiments on two datasets. The first benchmark set is used to illustrate the contribution of both the GA and the DBH. The second set is used to compare the results with current state-of-the-art heuristics, and to show that the procedure is capable of producing consistently good results for challenging instances of the resource-constrained project scheduling problem. We illustrate that GA is currently the best performing RCPSP meta-heuristic, and that the DBH further improves the performance of the GA

A Bi-Population Based Genetic Algorithm for the Resource-Constrained Project Scheduling Problem

Abstract: The resource-constrained project scheduling problem (RCPSP) is one of the most challenging problems in project scheduling. During the last couple of years many heuristic procedures have been developed for this problem, but still these procedures often fail in finding near-optimal solutions for more challenging problem instances. In this paper, we present a new genetic algorithm (GA) that, in contrast of a conventional GA, makes use of two separate populations. This bi-population genetic algorithm (BPGA) operates on both a population of left-justified schedules and a population of right-justified schedules in order to fully exploit the features of the iterative forward/backward local search scheduling technique. Comparative computational results reveal that this procedure can be considered as today’s best performing RCPSP heuristic. Note

The electromagnetism meta-heuristic applied to the resource-constrained project scheduling problem

Abstract: Recently, an electromagnetism (EM) heuristic has been introduced by Birbil and Fang (2003) to solve unconstrained optimization problems. In this paper, we extend the EM methodology to combinatorial optimization problems and illustrate its effectiveness on the well-known resource-constrained project scheduling problem (RCPSP). We present computational experiments on a standard benchmark dataset, compare the results of the different modifications on the original EM framework with current state-of-the-art heuristics, and show that the procedure is capable of producing consistently good results for challenging instances of the problem under study. We also give directions for future research in order to further explore the potential of this new technique.

Characterisation and Generation of Nurse Scheduling Problem Instances

Abstract: In this paper, we propose different complexity indicators for the well-known nurse scheduling problem (NSP). The NSP assigns nurses to shifts per day taking both hard and soft constraints into account. The objective is to maximize the nurses’ preferences and to minimize the total penalty cost from violations of the soft constraints. The problem is known to be NP-hard. Due to its complexity and relevance in practice, the operations research literature has been overwhelmed by different procedures to solve the problem. The complexity has resulted in the development of several (meta-)heuristic procedures, able to solve a NSP instance heuristically in an acceptable time limit. The practical relevance has resulted in a never-ending amount of different NSP versions, taking practical, case-specific constraints into account. The contribution of this paper is threefold. First, we describe our complexity indicators to characterize a nurse scheduling problem instance. Secondly, we develop a NSP generator to generate benchmark instances to facilitate the evaluation of existing and future research techniques. Finally, we perform some preliminary tests on a simple IP model to illustrate that the proposed indicators can be used as predictors of problem complexity.

A simulation and evaluation of earned value metrics to forecast the project duration

Abstract: It is well-known that well managed and controlled projects are more likely to be delivered on time and within budget The construction of a (resource-feasible) baseline schedule and the follow-up during execution are primary contributors to the success or failure of a project Earned value management systems have been set up to deal with the complex task of controlling and adjusting the baseline project schedule during execution Although earned value systems have been proven to provide reliable estimates for the follow-up of cost performance, it often fails to predict the total duration of the project In this paper, we extensively review the existing methods to forecast the total project duration Moreover, we investigate the potential of a newly developed method, the earned schedule method, which makes the connection between earned value metrics and the project schedule We present an extensive simulation study where we carefully control the level of uncertainty in the project, the influence of the project network structure on the accuracy of the forecasts, and the time horizon where the newly developed measures provide accurate and reliable results

A simulation analysis of interactions between errors in costing system design

Abstract: The academic accounting literature has established that the conditions under which costing systems in general and Activity Based Costing (ABC) in particular provide accurate costs are very stringent. Less is known, however, about the nature, level and bias of costing errors and their interactions, when these conditions are not met. The main problem to overcome to enable us to learn about these is the notion of the unobservable true cost benchmark to which to compare the costing system approximation. This paper presents a simulation method to deal with this problem, allowing a variety of research questions in this research area to be addressed with more generalizable answers. Using our methodology, we test a variety of hypotheses on the interaction between various errors in costing system design that were developed in the previous analytical, empirical, and practitioner literature. We also provide some interesting new insights on interactions between errors that were previously not discussed in the literature. This paper presents new results on (1) conditions under which partial refinement in costing systems does or does not work to improve overall accuracy, (2) the contexts in which it is most effective to correct a particular type of error in terms of improving overall accuracy and (3) indicators of robustness or sensitivity of costing system designs to errors. In doing so, we also provide insights relevant to practitioners, costing system designers and users of costing information alike.

A comparison of different project duration forecasting methods using earned value metrics

Abstract: Earned value project management is a well-known management system that integrates cost, schedule and technical performance. It allows the calculation of cost and schedule variances and performance indices and forecasts of project cost and schedule duration. The earned value method provides early indications of project performance to highlight the need for eventual corrective action. Earned value management was originally developed for cost management and has not widely been used for forecasting project duration. However, recent research trends show an increase of interest to use performance indicators for predicting total project duration. In this paper, we give an overview of the state-of-the-art knowledge for this new research trend to bring clarity in the often confusing terminology. The purpose of this paper is three-fold. First, we compare the classic earned value performance indicators SV & SPI with the newly developed earned schedule performance indicators SV(t) & SPI(t). Next, we present a generic schedule forecasting formula applicable in different project situations and compare the three methods from literature to forecast total project duration. Finally, we illustrate the use of each method on a simple one activity example project and on real-life project data.

An Electromagnetic Meta-Heuristic for the Nurse Scheduling Problem

Abstract: In this paper, we present a novel meta-heuristic technique for the nurse scheduling problem (NSP) This well-known scheduling problem assigns nurses to shifts per day taking both hard and soft constraints into account The objective is to maximize the preferences of the nurses and to minimize the total penalty cost from violations of the soft constraints The problem is known to be NP-hard Due to its complexity and relevance, many algorithms have been developed to solve practical, and often case-specific versions of the NSP The enormous amount of different constraints has led to an overwhelming amount of exact and meta-heuristic procedures, and hence comparison and stateof- the-art reporting of standard results seem to be a utopian idea The contribution of this paper is twofold First, we present a meta-heuristic procedure for the NSP based on the framework proposed by Birbil and Fang (2003) The Electromagnetic (EM) approach is based on the theory of physics, and simulates attraction and repulsion of sample points in order to move towards a promising solution Second, we present computational experiments on a standard benchmark dataset, and solve problem instances under different assumptions We show that our procedure performs consistently well under many different circumstances, and hence, can be considered as robust against case-specific constraints