A teaching–learning-based optimization algorithm for the resource-constrained project scheduling problem

Abstract

In this paper, a recently introduced population-based metaheuristic known as teaching–learning-based optimization algorithm (TLBO) is used to find solution for the resource-constrained project scheduling problem (RCPSP). The RCPSP is considered in its basic form, wherein activities are non-preemptive in nature and their execution is possible in a single mode only. The scheduling objective chosen is minimizing the makespan or total project duration. TLBO algorithm in its original form employs two phases, namely the teacher phase and the learner phase to reach a global optimum solution. In order to increase the exploitation and exploration capabilities of the basic TLBO the concepts of elitism and mutation as used in genetic algorithm (GA) have been introduced. An activity list representation is used to represent a learner (solution) and to derive the schedule from this activity list a serial schedule generation scheme is used as a decoding procedure. It has been found after the computational experiment on a test problem from the literature that proposed TLBO gives results competitive to the other metaheuristics like GA and particle swarm optimization (PSO). In addition, it offers the inherent advantage of less parameter to tune and can, therefore, be used as an effective method to solve RCPSP and its other variants as well.