Coordination Scheduling Problem Of Single Machine Manufacturing And Delivery With Due Windows

Coordination is one of kernel concepts about supply chain management. With rapiddevelopment of the global economy, it is very important for enterpirse to reduce cost, improvecu’stomers service and respond to market quickly. So, how to perform coordinationmanagement among different business partners becomes an important problem, and has beena hot research field about supply chain management.This thesis focuses on modeling andsolving coordination scheduling problems of single machine manufactuirng and delivery withdue windows.(1)Considering a two-stage supply chain system composed of one manufacturer andmultiple customers, a due window coordinating scheduling problem of manufactuirng anddelivery with constraints of one delivery batch for the same customer is studied. The problemcan be descirbed as follows: a set of jobs with vairous sizes are processed on single machineand then delivered to customers by a capacity limited vehicle respectively to meet thecustomers’ due windows. Jobs delivered at one time, called a delivery batch, only belong toone customer. The objective is to determine job processing sequence and delivery batchesscheduling so as to minimize the sum of makespan and earliness/tardiness penalties.According to characteirstics of the problem, mathematical model is built; the particle swarmoptimization (PSO) algorithm and the cultural algorithm (CA) are proposed and stated indetail, including representation of solution，initialization of population, construction of fitnessfunction, update strategy and revising operator. Some calculating expeirments are performedwith PSO and CA algorithms on different problem instances,the calculating results arepresented and discussed.(2)Consideirng a two-stage supply chain system composed of one manufacturer andmultiple customers, a due window coordinating scheduling problem of manufactuirng anddelivery with constraints of one delivery batch for the different customers is studied. Theproblem can be described as follows; a set of jobs with various sizes are processed on singlemachine and then delivered to customers by a capacity limited vehicle respectively to meet th’e customers due windows. Jobs delivered at one time, called a delivery batch, belong todifferent customers. The objective is to determine job processing sequence, delivery batchesand delivery paths scheduling so as to minimize the sum of makespan and. earliness/tardinesspenalties. According to characteirstics of the problem, mathematical model isbuilt.Consideirng whether there is a direct path connected between any two customers,namely:the different situations which the path can constitute a tirangle and the path can not constitutea tirangle, the particle swarm optimization (PSO) algorithm and the cultural algorithm (CA)are proposed respectively and stated in detail, including representation of solution,initialization of population, construction of fitness function, update strategy and revisingoperator.Some calculating expeirments are performed with PSO and CA algorithms ondifferent problem instances?the calculating results are presented and discussed.