A Multi-objective Genetic Algorithm For Optimizing Multiple Labor Shifts In Construction Projects

Multiple shifts are often used in construction projects to increase daily work hours and accelerate schedules. This utilization of evening or night shift often leads to negative impacts on construction cost and productivity. The shift workers who often struggle with adapting to new sleep cycles would confront more risk during working. Accelerating schedules while minimizing the cost and negative impacts of utilizing multiple shifts are expected by engineering managers.This paper presents a multi-objective optimization model for scheduling multiple labor shifts in construction projects, which searches for and identifies optimal/near optimal tradeoffs among minimizing project duration, reducing cost, and minimizing labor utilization in evening and night shifts. Two methods, multi-objective and single objective and Multi-subpopulations and single objective are provided in the chapters of implementation model. Niche genetic algorithm is used to explore diversity of population while Elitist selection strategy is adopted to reserve non-dominated solution.We present three experiments,including the experiment without consideration of optimizing the number of staff, the experiment with consideration of optimizing the number of staff and an improved genetic algorithm for optimizing multiple labor shifts.The result of comparison among these three experiments shows that the search range of the 3^rd experiment which adopts improved genetic algorithm is larger than other experiments. The 3^rd experiment can provide more reasonable solutions for engineering managers.