Some metaheuristic algorithms for solving multiple cross-functional team selection problems

Son Tung Ngo; Jafreezal Jaafar; Aziz Abdul Izzatdin; Giang Truong Tong; Anh Ngoc Bui

doi:10.7717/peerj-cs.1063

Some metaheuristic algorithms for solving multiple cross-functional team selection problems

PeerJ Comput Sci. 2022 Aug 9:8:e1063. doi: 10.7717/peerj-cs.1063. eCollection 2022.

Authors

Son Tung Ngo^{1

2}, Jafreezal Jaafar¹, Aziz Abdul Izzatdin¹, Giang Truong Tong², Anh Ngoc Bui²

Affiliations

¹ Department of Computer and Information Sciences, Universiti Teknologi PETRONAS, Seri Iskandar, Perak, Malaysia.
² Information and Communication Department, FPT University, Hà Noi, Vietnam.

Abstract

We can find solutions to the team selection problem in many different areas. The problem solver needs to scan across a large array of available solutions during their search. This problem belongs to a class of combinatorial and NP-Hard problems that requires an efficient search algorithm to maintain the quality of solutions and a reasonable execution time. The team selection problem has become more complicated in order to achieve multiple goals in its decision-making process. This study introduces a multiple cross-functional team (CFT) selection model with different skill requirements for candidates who meet the maximum required skills in both deep and wide aspects. We introduced a method that combines a compromise programming (CP) approach and metaheuristic algorithms, including the genetic algorithm (GA) and ant colony optimization (ACO), to solve the proposed optimization problem. We compared the developed algorithms with the MIQP-CPLEX solver on 500 programming contestants with 37 skills and several randomized distribution datasets. Our experimental results show that the proposed algorithms outperformed CPLEX across several assessment aspects, including solution quality and execution time. The developed method also demonstrated the effectiveness of the multi-criteria decision-making process when compared with the multi-objective evolutionary algorithm (MOEA).

Keywords: Ant colony optimization; CPLEX-MIQP; Combinatorial optimization; Compromise programming; Genetic algorithm; Multi objective optimization; Team selection.

Grants and funding

This research was funded by FPT University, under decision number QD1097/QD-DHFPT and QD1393/QD-DHFPT for Project HO-CPDT2021. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.