面向三次分配问题的高效启发式算法设计-软件工程专业毕业论文.docxVIP

面向三次分配问题的高效启发式算法设计The 面向三次分配问题的高效启发式算法设计 The High—Efficient Heuristic Algorithm Design for AP3 Problem Abstract 111e three—index assignment problem(also known as AP3 problem)is a type of classic combinatorial optimization problem．It is the extension of the linear assignment problem．111e aim of a linear assignment problem iS to do assignment on two finite sets，which have the same cardinality．111e linear assignment problem can be solved、析m the Hungarian Algorithm in polynomial time．The aim of a three-index assignment problem is to do assignment on three fimte sets．However．the three．index assignment problem has been proved as a NP—hard problem，thus，there is no polynomial time algorithm to find the optimal solution of the three-index assignment problem．The three—index assignment problem has wide applications， thus there have been lots of heuristics proposed to solve the three—index assignment problem． In this paper，a new heuristic named Approximate Muscle Guided Beam Search is purposed to solve the three．index assignment problem．This heuristic iS based on an effective strategy for NP—hard problem．the muscle(the union of optimal solutions)．T11is new heuristic analyses the instance firstly，to generate the approximate muscle for the instance to reduce the search space，and then it uses beam search to find the solution on the approximate muscle．By using muscle and beam search to reduce the search space，this heuristic achieves a good trade-off between running time and solution quality．The experiment on the benchmark shows that．the new heuristic is able to obtain a solution wim high quality while it can be employed on instances、析th large scale．This heuristic is not only an effective algorithm for the three—index assignment problem，but also a promising method for improving beam search． For the three-index assignment problem埘m large scale，in this paper，a new hyper-heuristic based on the common mathematical programming solver is purposed．The common mathematical programming solver