A Portfolio Approach to Algorithm Selection for Discrete Time-Cost Trade-off Problem

It has been widely known that performance of algorithms for NP-Hard problems varies from instance to instance. This phenomenon has been observed, when we comprehensively studied multi-objective evolutionary algorithms (MOEAs) on a six benchmark instances of discrete time-cost trade-off problem (DTCTP). Instead of using single algorithm to solve DTCTP, we use a portfolio approach that takes multiple algorithms as its constituent. In this paper, we proposed portfolio comprising of four MOEAs, Non-dominated sorting genetic algorithm 2 (NSGA 2), the strength Pareto EA 2 (SPEA 2), Pareto archive evolutionary strategy (PAES) and Niched Pareto Genetic Algorithm 2 (NPGA 2) to solve DTCTP. The result shows that the portfolio approach is computationally fast and qualitatively superior than its constituent algorithms for all benchmark instances. Moreover, portfolio approach provides an insight in selecting the best algorithm for all instances of DTCTP.