Sophie Demassey

A Cost-Regular Based Hybrid Column Generation Approach

Constraint Programming (CP) offers a rich modeling language of constraints embedding efficient algorithms to handle complex and heterogeneous combinatorial problems. To solve hard combinatorial optimization problems using CP alone or hybrid CP-ILP decomposition methods, costs also have to be taken into account within the propagation process. Optimization constraints, with their cost-based filtering algorithms, aim to apply inference based on optimality rather than feasibility. This paper introduces a new optimization constraint, cost-regular. Its filtering algorithm is based on the computation of shortest and longest paths in a layered directed graph. The support information is also used to guide the search for solutions. We believe this constraint to be particularly useful in modeling and solving Column Generation subproblems and evaluate its behaviour on complex Employee Timetabling Problems through a flexible CP-based column generation approach. Computational results on generated benchmark sets and on a complex real-world instance are given.

Global Constraint Catalogue: Past, Present and Future

The catalogue of global constraints is reviewed, focusing on the graph-based description of global constraints. A number of possible enhancements are proposed as well as several research paths for the development of the area.

Constraint-Propagation-Based Cutting Planes: An Application to the Resource-Constrained Project Scheduling Problem

We propose a cooperation method between constraint programming and integer programming to compute lower bounds for the resource-constrained project scheduling problem (RCPSP). The lower bounds are evaluated through linear-programming (LP) relaxations of two different integer linear formulations. Efficient resource-constraint propagation algorithms serve as a preprocessing technique for these relaxations. The originality of our approach is to use additionally some deductions performed by constraint propagation, and particularly by the shaving technique, to derive new cutting planes that strengthen the linear programs. Such new valid linear inequalities are given in this paper, as well as a computational analysis of our approach. Through this analysis, we also compare the two considered linear formulations for the RCPSP and confirm the efficiency of lower bounds computed in a destructive way.

Sequencing and Counting with the multicost-regular Constraint

This paper introduces a global constraint encapsulating a regular constraint together with several cumulative costs. It is motivated in the context of personnel scheduling problems, where a schedule meets patterns and occurrence requirements which are intricately bound. The optimization problem underlying the multicost-regular constraint is NP-hard but it admits an efficient Lagrangian relaxation. Hence, we propose a filtering based on this relaxation. The expressiveness and the efficiency of this new constraint is experimented on personnel scheduling benchmark instances with standard work regulations. The comparative empirical results show how multicost-regular can significantly outperform a decomposed model with regular and global-cardinality constraints.

Constraint programming based column generation for employee timetabling

The Employee Timetabling Problem (ETP) is a general class of problems widely encountered in service organizations (such as call centers for instance). Given a set of activities, a set of demand curves (specifying the demand in terms of employees for each activity for each time period) the problem consists of constructing a set of work shifts such that each activity is at all time covered by a sufficient number of employees. Work shifts can cover many activities and must meet work regulations such as breaks, meals and maximum working time constraints. Furthermore, it is often desired to optimize a global objective function such as minimizing labor costs or maximizing a quality of service measure. This paper presents variants of this problem which are modeled with the Dantzig formulation. This approach consists of first generating all feasible work shifts and then selecting the optimal set. We propose to address the shift generation problem with constraint satisfaction techniques based on expressive and efficient global constraints such as gcc and regular. The selection problem, which is modeled with an integer linear program, is solved by a standard MIP solver for smaller instances and addressed by column generation for larger ones. Since a column generation procedure needs to generate only shifts of negative reduced cost, the optimization constraint cost-regular is introduced and described. Preliminary experimental results are given on a typical ETP.

Graph properties based filtering

This article presents a generic filtering scheme, based on the graph description of global constraints. This description is defined by a network of binary constraints and a list of elementary graph properties: each solution of the global constraint corresponds to a subgraph of the initial network, retaining only the satisfied binary constraints, and which fulfills all the graph properties. The graph-based filtering identifies the arcs of the network that belong or not to the solution subgraphs. The objective is to build, besides a catalog of global constraints, also a list of systematic filtering rules based on a limited set of graph properties. We illustrate this principle on some common graph properties and provide computational experiments of the effective filtering on the group constraint.

Lower Bounds for Resource Constrained Project Scheduling Problem

We review the most recent lower bounds for the makespan minimization variant of the Resource Constrained Project Scheduling Problem. Lower bounds are either based on straight relaxations of the problems (e.g., single machine, parallel machine relaxations) or on constraint programming and/or linear programming formulations of the problem.