Optimization Method based on Genetic Algorithms using Parameterized Active Schedules for Project Scheduling with Limited Resources

摘要

A project can be depicted by a graph where the activities are numerically numbered. Associated with each activity is a set of possible durations with specific resource requirements. If resources are available in limited quantities each time period, the resources are considered renewable (e.g., machines or manpower). As the number of project activities increases and thus the complexity of their sequential ordering, the need for organized planning and scheduling increases too. This need further increases when a large number of project activities are considered relative to the uniqueness of each construction project in terms of the dynamic plant and nonstandardized nature of the work. So, finding feasible schedules which efficiently use scarce resources is a challenging task within project management. This type of problem belongs to the class of NP-hard optimization problems, therefore justifying the indispensable use of heuristic or metaheuristic solution procedures when solving large problem instances. The optimization methods presented combines genetic algorithms and a schedule generator scheme which generates parameterized active schedules. The chromosome representation of the problem is based on random keys. Parameterized active schedules are constructed using a priority-rule heuristic in which the priorities of the activities are defined by the genetic algorithm.