We study questions of robustness of linear multiple objective problems in the sense of post-optimal analysis, that is, we study conditions under which a given efficient solution remains efficient when the criteria/objective matrix undergoes some alterations. We consider addition or removal of certain criteria, convex combination with another criteria matrix, or small perturbations of its entries. We provide a necessary and sufficient condition for robustness in a verifiable form and give two formulae to compute the radius of robustness.
В статье рассматриваются задачи выбора, в которых предпочтения лица, принимающего решение, представлены в форме параметрического слабого порядка без предположения о существовании функции ценности. Исследуется чувствительность (устойчивость) к изменению параметров этого порядка как каждого из недоминируемых решений, так и всего множества таких решений в целом. Показывается, что такой тип анализа чувствительности может быть проведен с использованием методов линейного программирования.
This paper investigates the coordinated scheduling problem of production and transportation in a two-stage supply chain, where the actual job processing time is a linear function of its starting time. During the production stage the jobs are first processed in serial batches on a bounded serial batching machine at the manufacturer's site. Then, the batches are delivered to a customer by a single vehicle with limited capacity during the transportation stage, and the vehicle can only deliver one batch at one time. The objective of this proposed scheduling problem is to make decisions on job batching and batch sequencing so as to minimize the makespan. Moreover, we consider two different models. With regards to the scheduling model with a buffer for storing the processed batches before transportation, we develop an optimal algorithm to solve it. For the scheduling model without buffer, we present some useful properties and develop a heuristic H for solving it. Then a novel lower bound is derived and two optimal algorithms are designed for solving two special cases. Furthermore, computational experiments with random instances of different sizes are conducted to evaluate the proposed heuristic H, and the results show that our proposed algorithm is superior to other four approaches in the literature. Besides, heuristic H in our experiments can effectively and efficiently solve both small-size and large-size problems in a reasonable time.
The main goal of this paper is to model the effects of wholesale price control on manufacturer’s profit, taking explicitly into account the retailer’s sales motivation and performance. We consider a stylized distribution channel where a manufacturer sells a single kind of good to a single retailer. Wholesale price discounts are assumed to increase the retailer’s motivation thus improving sales. We study the manufacturer’s profit maximization problem as an optimal control model where the manufacturer’s control is the discount on wholesale price and retailer’s motivation is one of the state variables. In particular in the paper we prove that an increasing discount policy is optimal for the manufacturer when the retailer is not efficient while efficient retailers may require to decrease the trade discounts at the end of the selling period. Computational experiments point out how the discount on wholesale price passed by the retailer to the market (pass-through) influences the optimal profit of the manufacturer.