带有恶化率的可拒绝单机排序及批配送

摘要

In this paper,the problem of scheduling linear deteriorating jobs with rejection on a single machine and batch delivery is considered.A job is either rejected,in which case a rejection penalty has to be paid,or accepted and processed on the machine.The obj ective is to minimize the sum of the total weighted completion time(or the maximum lateness)of the accepted jobs,the transportation costs,and the total re-jection penalty of the rejected jobs.These problems are proved to be NP-hard in the ordinary sense,and pseudo-polynomial-time dynamic programming algorithms is proposed.%首次考虑了加工时间带有线性恶化率的可拒绝单机排序及其批配送的问题。如果工件被拒绝，则要付出一定的拒绝费用；如果工件被接受，则要安排加工并配送。目标函数是极小化接受工件的加权总完工时间或最大延误时间，配送费用与拒绝工件的拒绝费用这三部分的和，我们不仅证明了这些问题都是 NP-hard的，而且还提出了基于动态规划的伪多项式时间算法。