Integer Programming to Minimize Labour Costs
本文展示了纯整数规划在劳动力成本问题中的实际应用，通过分析不同班次策略，找到满足预测需求的最小成本方案，并发现舍入连续最优解得到的整数解并非最优。
AbstractThe main purpose of this paper is to demonstrate a real-world application of pure integer programming to find the optimum solution to a labour cost problem. The length of a daily working shift is defined as an integer variable and several shift strategies are analysed to determine the optimum length and shift combinations that satisfy a predicted demand at minimum cost. The state-space model has been used to predict the stochastic behaviour of monthly demands for beer and soft drink. Savings of about 7% of the annual sales have been obtained as a result of implementing the integer programming approach. A numerical example shows that the solution obtained by rounding off the continuous optimal solution does not match with the integer optimal solution. It was also noted that if a rounded-off solution is feasible, then it provides an initial integer solution for the branch-and-bound algorithm that may reduce the computational time.Keywords: integer programminglabour costsstate-space model