Using non-dominated sorting-II teaching learning-based optimization (NDSII-TLBO) in solving time-cost-quality trade-off problems

In today's construction industry, all parties strive to complete the project in the shortest amount of time, at the lowest possible cost, and with the highest possible quality. There has been a lot of research done on multi-objective optimization in construction projects. Initially, studies focused on time-cost trade-off, but recent efforts incorporate more objectives such as safety, quality, resources, environment, sustainability, etc., in traditional time-cost tradeoff. This study considers multi-objective functions to minimize total project time and total cost while maximizing overall project quality. Teaching Learning-Based Optimization (TLBO) is a population-based optimization algorithm inspired by the transmission of information in the classroom; Non-dominated sorting-II (NDS-II) is used to sort population solutions according to the Pareto dominance principle, which is crucial in the selection operation of many multi-objective evolutionary algorithms; and crowding distance measure describes the excellence of the solutions in the same rank. In this study, the NDS-II concept and the mechanism of crowding distance computation are incorporated with the TLBO algorithm to optimize time-cost-quality optimization problems. The incorporated model is coded in MATLAB and applied to four different time-cost-quality case studies sized from 7 to 18 activities. Keywords: Time-cost-quality trade-off problems, meta-heuristic methods, teaching learning-based optimization (TLBO), non-dominating sorting-II (NDS-II)