Industrial petrochemical production planning and expansion: A multi-objective linear programming approach

This paper considers a strategic model planning for the petrochemical industry. It concerns with the expansion in a firm producing multiple products in several regions of a country. The expansion of the existing facilities and the new ones are considered. It also exists a large amount of interdependencies among the firm’s products, because the output of one particular plant can be used as an input to the production of another plant in the same or different regions and to satisfy the final demand. The decision makers involved in the planning process should identify several objectives. Then, multiple objective programming is used for making trade-offs among the economic and operational factors considered. To define the interval criteria weights into the model we utilized the Analytic Hierarchy Process to bring them closer to the decision makers preferences. This work was sponsored by the Institut National Polytechnique de Toulouse, France, when the author was Associate Professor at the Département Génie des Systèmes Industriels.     

Integration of generative and evaluative models for production scheduling of lube oil plants in a petroleum refinery

A study was made on the existing practices of production planning, scheduling and prevailing constraints in the six plants of a lube oil section in a petroleum refinery. Based on the data collected from these plants, some generative and evaluative models were developed. The generative models developed were flow network optimisation (FNO) model and binary integer linear programming (BILP) model. The evaluative model developed was simulation. The optimal results obtained from the generative model were fed to the evaluative model to derive the measure of performance. This integration of generative and evaluative models offers an opportunity for better understanding of the subsystem and appropriate decision making.     

An interactive,computer-aided ship scheduling system

This paper is concerned with a fleet scheduling and inventory resupply problem faced by an international chemical operation. The firm uses a fleet of small ocean-going tankers to deliver bulk fluid to warehouses all over the world. The scheduling problem centers around decisions on routes, arrival/departure times, and inventory replenishment quantities. An interactive computer system was developed and implemented at the firm, and was successfully used to address daily scheduling issues as well as longer range planning problems. The purpose of this paper is to first present how the underlying decision problem was analyzed using both a network flow model and a mixed integer programming model, and then to describe the components of the decision support system developed to generate schedules. The use of the system in various decision making applications is also described.     

Planning production in a bottleneck department

This paper discusses the development and the implementation of a production planning model for a set of bottleneck machines. The machines are characterized by very high start/stop costs while demand for the products is time-varying and deterministic. Setup times are highly sequence-dependent. The model uses Lagrangian relaxation and dynamic programming to determine the production periods for the machines. Capacitated lot sizing models are then used to establish individual product runs on each machine. Finally, the actual sequences are found by applying a traveling salesman algorithm. Though the model is presented in some detail, emphasis will be on how the model was developed and implemented in close cooperation with the executives of the firm and on how the model is currently used for decision making.     

A simple Tabu Search method to solve the mixed-integer linear bilevel programming problem

Multilevel programming is characterized as mathematical programming to solve decentralized planning problems. The models partition control over decision variables among ordered levels within a hierarchical planning structure of which the linear bilevel form is a special case of a multilevel programming problem. In a system with such a hierarchical structure, the high-level decision making situations generally require inclusion of zero-one variables representing ‘yes-no’ decisions. We provide a mixed-integer linear bilevel programming formulation in which zero-one decision variables are controlled by a high-level decision maker and real-value decision variables are controlled by a low-level decision maker. An algorithm based on the short term memory component of Tabu Search, called Simple Tabu Search, is developed to solve the problem, and two supplementary procedures are proposed that provide variations of the algorithm. Computational results disclose that our approach is effective in terms of both solution quality and efficiency.     

Robust optimization for interactive multiobjective programming with imprecise information applied to R&D project portfolio selection

A multiobjective binary integer programming model for R&D project portfolio selection with competing objectives is developed when problem coefficients in both objective functions and constraints are uncertain. Robust optimization is used in dealing with uncertainty while an interactive procedure is used in making tradeoffs among the multiple objectives. Robust nondominated solutions are generated by solving the linearized counterpart of the robust augmented weighted Tchebycheff programs. A decision maker’s most preferred solution is identified in the interactive robust weighted Tchebycheff procedure by progressively eliciting and incorporating the decision maker’s preference information into the solution process. An example is presented to illustrate the solution approach and performance. The developed approach can also be applied to general multiobjective mixed integer programming problems.     

基于模糊需求的多产品供应商选择的多目标规划模型

由决策于环境的不确定性,供应商选择问题存在大量的模糊信息,传统的确定性规划模型已经不能够很好地处理此类问题。本文基于模糊需求量信息,对于多产品供应商问题建立了模糊多目标规划模型。同时考虑到各目标及约束的重要性程度不同的影响,通过引进适当的权重对多目标规划模型进行求解。文中结合实际算例验证模型的可行性和有效性。     

Development of a new approach for deterministic supply chain network design

This paper proposes a mixed integer linear programming model and solution algorithm for solving supply chain network design problems in deterministic, multi-commodity, single-period contexts. The strategic level of supply chain planning and tactical level planning of supply chain are aggregated to propose an integrated model. The model integrates location and capacity choices for suppliers, plants and warehouses selection, product range assignment and production flows. The open-or-close decisions for the facilities are binary decision variables and the production and transportation flow decisions are continuous decision variables. Consequently, this problem is a binary mixed integer linear programming problem. In this paper, a modified version of Benders’ decomposition is proposed to solve the model. The most difficulty associated with the Benders’ decomposition is the solution of master problem, as in many real-life problems the model will be NP-hard and very time consuming. In the proposed procedure, the master problem will be developed using the surrogate constraints. We show that the main constraints of the master problem can be replaced by the strongest surrogate constraint. The generated problem with the strongest surrogate constraint is a valid relaxation of the main problem. Furthermore, a near-optimal initial solution is generated for a reduction in the number of iterations.     

The construction of a comprehensive model for production strategy evaluation

To acquire reasonable profits and to be competitive in the globalize market, more and more firms today have exploited various kinds of production strategies, such as outsourcing, joint venture, or some kind of strategic alliance. Since every production strategy has its benefits and costs and may bring a firm different opportunities and risks, which kind of production strategy is the most suitable for a firm to carry out is a difficult and complicated decision with a high degree of uncertainty. Therefore, this research proposes a comprehensive production strategy evaluation model that can facilitate such a decision making for a firm. The factors that should be considered for devising a production strategy are listed for the benefits, opportunities, costs and risks (BOCR) merits first, and fuzzy Delphi method is applied next to select the most important factors under each merit. A network with BOCR merits is constructed to incorporate the selected factors, and fuzzy analytic network process is then applied to consider the interdependence and feedback inside the network. The proposed model can effectively aid decision making on which kind of production strategy should be adopted by a firm. A case of a printed circuit board manufacturer is presented for the illustration of the proposed model.     

Equity portfolio construction and selection using multiobjective mathematical programming

A multi-objective mixed integer programming model for equity portfolio construction and selection is developed in this study, in order to generate the Pareto optimal portfolios, using a novel version of the well known ε-constraint method. Subsequently, an interactive filtering process is also proposed to assist the decision maker in making his/her final choice among the Pareto solutions. The proposed methodology is tested through an application in the Athens Stock Exchange.     

On computational methods for solutions of multiobjective linear production programming games

In this paper we consider a production model in which multiple decision makers pool resources to produce finished goods. Such a production model, which is assumed to be linear, can be formulated as a multiobjective linear programming problem. It is shown that a multi-commodity game arises from the multiobjective linear production programming problem with multiple decision makers and such a game is referred to as a multiobjective linear production programming game. The characteristic sets in the game can be obtained by finding the set of all the Pareto extreme points of the multiobjective programming problem. It is proven that the core of the game is not empty, and points in the core are computed by using the duality theory of multiobjective linear programming problems. Moreover, the least core and the nucleolus of the game are examined. Finally, we consider a situation that decision makers first optimize their multiobjective linear production programming problem and then they examine allocation of profits and/or costs. Computational methods are developed and illustrative numerical examples are given.     

A market-based multi-agent system model for decentralized multi-project scheduling

We consider a multi-project scheduling problem, where each project is composed of a set of activities, with precedence relations, requiring specific amounts of local and shared (among projects) resources. The aim is to complete all the project activities, satisfying precedence and resource constraints, and minimizing each project schedule length. The decision making process is supposed to be decentralized, with as many local decision makers as the projects. A multi-agent system model, and an iterative combinatorial auction mechanism for the agent coordination are proposed. We provide a dynamic programming formulation for the combinatorial auction problem, and heuristic algorithms for both the combinatorial auction and the bidding process. An experimental analysis on the whole multi-agent system model is discussed.     

装配型企业订货量分配与订单排产联合决策研究

针对供应商交货数量不确定环境下,多品种小批量装配型制造企业因生产物料不配套造成生产计划不可行甚至客户订单拖期的问题,从企业运作整体出发,考虑订货量分配决策对订单生产和交货的影响,以最小化采购成本和最小化订单排产相关成本为优化目标,在允许零部件拖期交货且供应商提供拖期价格折扣条件下,建立订货量分配与订单排产联合优化模型。针对可行解空间巨大、传统数学规划方法难以求解的问题,从增强搜索性能角度出发,设计基于自定义邻域搜索算子的局部搜索机制和基于随机与种群重构变异机制的改进粒子群算法的模型求解策略。通过应用实例对本文模型和算法进行了有效性验证和灵敏度分析,结果表明,相比于传统的分散决策方案,本文模型能够有效降低整体成本水平,引入的改进机制能够显著提升算法搜索性能,为企业供应风险下的运营决策制定提供理论参考。     

A stochastic programming approach for multi-site aggregate production planning

Production planning problems play a vital role in the supply chain management area, by which decision makers can determine the production loading plan—consisting of the quantity of production and the workforce level at each production plant—to fulfil market demand. This paper addresses the production planning problem with additional constraints, such as production plant preference selection. To deal with the uncertain demand data, a stochastic programming approach is proposed to determine optimal medium-term production loading plans under an uncertain environment. A set of data from a multinational lingerie company in Hong Kong is used to demonstrate the robustness and effectiveness of the proposed model. An analysis of the probability distribution of economic demand assumptions is performed. The impact of unit shortage costs on the total cost is also analysed.     

An application of multi-objective modeling: The case of the Indian sugar industry

A multi-objective linear programming model is developed for the Indian sugar industry to plan for additional output by production technique, geographical region and forecasted year. Various policy scenarios generated by assigning different values to the policy variables in the model are studied. Thus a useful planning tool which demonstrates the exact impact of the policy parameters on various objectives is provided to the central decision maker. A satisficing multi-objective decision making method is developed based on an existing method of solution and used in policy analysis. The solution method is ideally suited to any general planning problem.     

Competitive facility location on decentralized supply chains

This paper addresses a novel competitive facility location problem about a firm that intends to enter an existing decentralized supply chain comprised of three tiers of players with competition: manufacturers, retailers and consumers. It first proposes a variational inequality for the supply chain network equilibrium model with production capacity constraints, and then employs the logarithmic-quadratic proximal prediction–correction method as a solution algorithm. Based on this model, this paper develops a generic mathematical program with equilibrium constraints for the competitive facility location problem, which can simultaneously determine facility locations of the entering firm and the production levels of these facilities so as to optimize an objective. Subsequently, a hybrid genetic algorithm that incorporates with the logarithmic-quadratic proximal prediction–correction method is developed for solving the proposed mathematical program with an equilibrium constraint. Finally, this paper carries out some numerical examples to evaluate proposed models and solution algorithms.     

A Linear Goal Programming Model for Trans-Shipment Problems with Flexible Supply and Demand Constraints

This paper deals with the development of a linear goal programming model for solving transshipment problems with flexible supply and demand constraints. The model was constructed based on the data obtained from a nationwide retail firm. The Wilcoxon signed-ranked test was used to show the model's flexibility and applicability. The test results were analysed and interpreted for decision making.     

A hybrid of genetic algorithm and particle swarm optimization for solving bi-level linear programming problem – A case study on supply chain model

The main goal of supply chain management is to coordinate and collaborate the supply chain partners seamlessly. On the other hand, bi-level linear programming is a technique for modeling decentralized decision. It consists of the upper level and lower level objectives. Thus, this paper intends to apply bi-level linear programming to supply chain distribution problem and develop an efficient method based on hybrid of genetic algorithm (GA) and particle swarm optimization (PSO). The performance of the proposed method is ascertained by comparing the results with GA and PSO using four problems in the literature and a supply chain distribution model.     

Short-run and long-run efficiency measures for multiplant firms

The usual Data Envelopment Analysis (DEA) model for measuring the relative efficiency assumes that all plants belong to distinct firms superior to them. For firms with more than one plant, Koopmans proposes a procedure for deriving the short-run production frontier for each firm. Modifying his idea, a DEA model is constructed in this paper for measuring the short-run efficiency of each plant within a firm. Based on the theory of production economics that the long-run production frontier is an envelop super-imposed upon all short-run production frontiers, another DEA model is constructed to measure the long-run efficiency of every plant. The long-run efficiency is always smaller than or equal to the short-run efficiency. Consequently, it is possible that an inefficient plant can only be improved in the long-run. With the models constructed in this paper, a decision-maker is able to distinguish between what can be achieved in the short-run and what in the long-run. To clarify the idea, an example of Taiwan forests is adopted for illustration. This revised version was published online in June 2006 with corrections to the Cover Date.     

Impact of productivity on cross-training configurations and optimal staffing decisions in hospitals

Cross-training of nursing staff has been used in hospitals to reduce labor cost, provide scheduling flexibility, and meet patient demand effectively. However, cross-trained nurses may not be as productive as regular nurses in carrying out their tasks because of a new work environment and unfamiliar protocols in the new unit. This leads to the research question: What is the impact of productivity on optimal staffing decisions (both regular and cross-trained) in a two-unit and multi-unit system. We investigate the effect of mean demand, cross-training cost, contract nurse cost, and productivity, on a two-unit, full-flexibility configuration and a three-unit, partial flexibility and chaining (minimal complete chain) configurations under centralized and decentralized decision making. Under centralized decision making, the optimal staffing and cross-training levels are determined simultaneously, while under decentralized decision making, the optimal staffing levels are determined without any knowledge of future cross-training programs. We use two-stage stochastic programming to derive closed form equations and determine the optimal number of cross-trained nurses for two units facing stochastic demand following general, continuous distributions. We find that there exists a productivity level (threshold) beyond which the optimal number of cross-trained nurses declines, as fewer cross-trained nurses are sufficient to obtain the benefit of staffing flexibility. When we account for productivity variations, chaining configuration provides on average 1.20% cost savings over partial flexibility configuration, while centralized decision making averages 1.13% cost savings over decentralized decision making.