Abstract: For the Small and Medium Production Enterprises(SMPEs) under MTO(Make-to-order) environment, materials provision in time and schedules as planed are the key factors, which affect delivery time. In reality, the price of some materials fluctuates over time continuously, and because of other random inserted orders, staff turnover, equipment failures and rework owing to the poor quality of some products etc, the makespan is random. For the decision of materials ordering, the traditional methods mainly consider the beginning date of the order and minimizing the materials purchasing costs but regardless of the random makespan. Based on the system considering about internal supply chain integration and minimizing the total costs, When the materials price fluctuates over time and the makespan is random, it considers how to model, analyze and solve the problem of getting the optimal decision of ordering date, so that the total costs (it includes the costs of materials procurement, manufacturing, delay in delivery or in advance) are lowest. For the case of a single type of materials, the existence of the minimum points by extreme method is discussed.When the types are more than two, a constraint nonlinear optimization model with multi-variables is built by the algorithm of PHR. Then, the numerical examples show the validity of the models. The above ideas about integrated procurement and production, the methods about model building and problems solving would give references to some relevant persons.

Key words: MTO, prices fluctuation, random makespan, ordering decisions, algorithm of PHR