The Basic Multi-Project Scheduling Problem

Abstract

In this chapter the Basic Multi-Project Scheduling Problem (BMPSP) is described, an overview of the literature on multi-project scheduling is provided, and a solution approach based on a biased random-key genetic algorithm (BRKGA) is presented. The BMPSP consists in finding a schedule for all the activities belonging to all the projects taking into account the precedence constraints and the availability of resources, while minimizing some measure of performance. The representation of the problem is based on random keys. The BRKGA generates priorities, delay times, and release dates, which are used by a heuristic decoder procedure to construct parameterized active schedules. The performance of the proposed approach is validated on a set of randomly generated problems. Keywords Genetic algorithm • Meta-heuristics • Multi-project scheduling • Random keys IntroductionManaging multiple projects is a complex decision-making process, where a number of projects must share concurrently a set of limited resources. Examples of multiproject environments are new product development, multi-product manufacturing, infrastructure constructions, and maintenance of systems.

Dynamic multi-mode resource-constrained multi-project scheduling problem with weighted earliness and tardiness: a real-life boutique furniture implementation

“…Project managers often deal with multiple projects simultaneously in real life and allocate scarce resources in a multi-project environment. Executing multiple projects together is a complex decisionmaking problem (Gonçalves, Mendes and Resende, 2015), and the multi-project scheduling problem is of the NP-hard type. Different modes can be defined for each activity with a different duration and resource requirement for each mode in some problems.…”