Energy supply planning and supply chain optimization under uncertainty

In energy supply planning and supply chain design, the coupling between long-term planning decisions like capital investment and short-term operation decisions like dispatching present a challenge, waiting to be tackled by systems and control engineers. The coupling is further complicated by uncertainties, which may arise from several sources including the market, politics, and technology. This paper addresses the coupling in the context of energy supply planning and supply chain design. We first discuss a simple two-stage stochastic program formulation that addresses optimization of an energy supply chain in the presence of uncertainties. The two-stage formulation can handle problems in which all design decisions are made up front and operating parameters act as ‘recourse’ decisions that can be varied from one time period to next based on realized values of uncertain parameters. The design of a biodiesel production network in the Southeastern region of the United States is used as an illustrative example. The discussion then moves on to a more complex multi-stage, multi-scale stochastic decision problem in which periodic investment/policy decisions are made on a time scale orders of magnitude slower than that of operating decisions. The problem of energy capacity planning is introduced as an example. In the particular problem we examine, annual acquisition of energy generation capacities of various types are coupled with hourly energy production and dispatch decisions. The increasing role of renewable sources like wind and solar necessitates the use of a fine-grained time scale for accurate assessment of their values. Use of storage intended to overcome the limitations of intermittent sources puts further demand on the modeling and optimization. Numerical challenges that arise from the multi-scale nature and uncertainties are reviewed and some possible modeling and numerical solution approaches are discussed.     

Comparing classical performance measures for a multi-period,two-echelon supply chain network design problem with sizing decisions

This paper addresses a new problem to design a two-echelon supply chain network over a multi-period horizon. Strategic decisions are subject to a given budget and concern the location of new facilities in the upper and intermediate echelons of the network as well as the installation of storage areas to handle different product families. A finite set of capacity levels for each product family is available at each potential location. Further decisions concern the quantities of products to be shipped through the network. Two mixed-integer linear programming models are proposed that differ in the type of performance measure that is adopted to design the supply chain. Under a cost minimization objective, the network configuration with the least total cost is to be determined. In contrast, under a profit maximization goal the aim is to design the network so as to maximize the difference between total revenue and total cost. In this case, it may not always be attractive to completely satisfy demand requirements. To investigate the implications that the choice of these performance measures have on network design, an extensive computational study is conducted with randomly generated instances that are solved using CPLEX.     

Design of robust cellular manufacturing system for dynamic part population considering multiple processing routes using genetic algorithm

In this paper, a comprehensive mathematical model is proposed for designing robust machine cells for dynamic part production. The proposed model incorporates machine cell configuration design problem bridged with the machines allocation problem, the dynamic production problem and the part routing problem. Multiple process plans for each part and alternatives process routes for each of those plans are considered. The design of robust cell configurations is based on the selected best part process route from user specified multiple process routes for each part type considering average product demand during the planning horizon. The dynamic part demand can be satisfied from internal production having limited capacity and/or through subcontracting part operation without affecting the machine cell configuration in successive period segments of the planning horizon. A genetic algorithm based heuristic is proposed to solve the model for minimization of the overall cost considering various manufacturing aspects such as production volume, multiple process route, machine capacity, material handling and subcontracting part operation.     

Freight transportation function in supply chain optimization models: A critical review of recent trends

Supply chain mathematical programming models mainly used for computer-aided decision-making processes, have been widely used to date as an advanced support to the experts’ opinions. Transportation operations are often a central aspect of such models. We developed a detailed review of the freight transportation function included in supply chain models, and some transportation aspects were identified and analyzed in recent articles (2009–2013). We identified one paradigm, two trends and an anomaly regarding transportation modeling. The main anomaly-related observation refers to the absence of correspondence between the modeling of transportation operations and the transportation cost function considered in the models. This gap has not been so far considered in the literature and we propose a framework to guide, in a more systemic way, the transportation considerations into optimization models. In addition, some concerns regarding trade-off analysis, private/outsourced fleet considerations, the role of time and distance in transportation cost analysis, among others, were also included. These issues are expected to be significant for supply chain analysts whose decisions emerge from modeling and computer-aided tools.     

Vertical collaboration in the semiconductor industry: A decision framework for supply chain relationships

Vertical collaboration problem focuses on integrating and modeling the decision problems of the suppliers and buyers together with the market intermediary by identifying the inefficiencies in the traditional marketplace and aligning the incentives of members in the e-marketplace. The present work develops and solves real life e-marketplace models for complex buyers–suppliers procurement problems by estimating the order quantities in the collaborated supply chain. The newsvendor framework considers demand to be independent of the selling price as is generally the case in the semiconductor industry supply chain dealing with techno-savvy customers. The vertical collaboration process would be more effective if the length of the planning horizon and order size is considered as a negotiation parameter between the buyer and supplier. It is observed that the supplier’s expected profit function increases with the buyers’ ordering quantity, which is important in characterizing the general structure of the collaboration scheme of the supply chain.     

A graph theoretic-based heuristic algorithm for responsive supply chain network design with direct and indirect shipment

The configuration of the supply chain network has a strong influence on the overall performance of the supply chain. A well designed supply chain network provides a proper platform for efficient and effective supply chain management. The supply chain network should be designed in the way that could meet the customer needs with an efficient cost. This paper studies the responsive, multi-stage supply chain network design (SCND) problem under two conditions: (1) when direct shipment is allowed and (2) when direct shipment is prohibited. First, two mixed integer programming models are proposed for multi-stage, responsive SCND problem under two abovementioned conditions. Then, to escape from the complexity of mixed integer mathematical programming models, graph theoretic approach is used to study the structure of the SCND problems and it is proven that both of SCND problems considered in this paper could be modeled by a bipartite graph. Finally, since such network design problems belong to the class of NP-hard problems, a novel heuristic solution method is developed based on a new solution representation method derived from graph theoretic view to the structure of the studied problem. To assess the performance of the proposed heuristic solution method, the associated results are compared to the exact solutions obtained by a commercial.     

A large neighborhood search heuristic for supply chain network design

Many exact and approximate solution techniques have been used to solve facility location problems and, more generally, supply chain network design problems. Yet, the Large Neighborhood Search technique (LNS) has almost never been suggested for solving such problems, although it has proven its efficiency and flexibility in solving other complex combinatorial optimization problems. In this paper, we propose an LNS framework for solving a four-layer single period multi-product supply chain network design problem. One important feature of the model is that it includes inter-modality: the itinerary followed by the cargo from origin to destination may take several transportation modes. Moreover, several modes may compete on some arcs. Location decisions for intermediate facilities (e.g. plants and distribution centers) are determined by the LNS while transportation modes and product flow decisions are determined by a greedy heuristic. As a post-optimization step, linear programming is used to optimize product flows once the structure of the logistics network is fixed. Extensive experiments, based on randomly generated instances of different sizes and characteristics, show the effectiveness of the method compared with a state-of-the-art solver.     

A particle swarm optimization algorithm for solving unbalanced supply chain planning problems

This paper focuses on developing a decision methodology for the production and distribution planning of a multi-echelon unbalanced supply chain. In the supply chain system discussed here, multiple products, production loss, transportation loss, quantity discount, production capacity, and starting-operation quantity are considered simultaneously, and the system pattern is ascertained with based on appropriate partners and suitable transportation quantities. To make a quality decision in supply chain planning, we first propose an optimization mathematical model which integrates cost and time criteria. Then, a particle swarm optimization (PSO) solving method is proposed for obtaining acceptable results is called MEDPSO. The MEDPSO introduces the maximum possible quantity strategy into the basic procedure of PSO to generate the initial feasible population in a timely fashion and provides an exchange and disturbance mechanism to prevent particle lapse into the local solution. Finally, one case and two simulated supply chain structures are proposed to illustrate the effectiveness of the MEDPSO method by comparing the results of classical GA and PSO in solving multi-echelon unbalanced supply chain planning problems with quantity discount.     

A dual-channel network design model in a green supply chain considering pricing and transportation mode choice

A green supply chain with a well-designed network can strongly influence the performance of supply chain and environment. The designed network should lead the supply chain to efficient and effective management to meet the efficient profit, sustainable effects on environment and customer needs. The proposed mathematical model in this paper identifies locations of productions and shipment quantity by exploiting the trade-off between costs, and emissions for a dual channel supply chain network. Due to considering different prices and customers zones for channels, determining the prices and strategic decision variables to meet the maximum profit for the proposed green supply chain is contemplated. In this paper, the transportation mode as a tactical decision has been considered that can affect the cost and emissions. Lead time and lost sales are considered in the modeling to reach more reality. The developed mathematical model is a mixed integer non-linear programming which is solved by GAMS. Due to NP-hard nature of the proposed model and long run time for large-size problems by GAMS, artificial immune system algorithm based on CLONALG, genetic and memetic algorithms are applied. Taguchi technique is used for parameter tuning of all meta-heuristic algorithms. Results demonstrate the strength of CLONALG rather than the other methods.     

A multi-objective optimization for green supply chain network design

In this paper, we study a supply chain network design problem with environmental concerns. We are interested in the environmental investments decisions in the design phase and propose a multi-objective optimization model that captures the trade-off between the total cost and the environment influence. We conduct a comprehensive set of numerical experiments. The results show that our model can be applied as an effective tool in the strategic planning for green supply chain. Meanwhile, the sensitivity analysis provides some interesting managerial insights for firms.     

Mathematical model for simultaneous design of module commonalization and supply chain configuration toward global product family

Manufacturing today has become global in all aspects marketing, design, production, distribution, etc. While product family design has been an essential viewpoint for meeting the demand for product variety, its interaction with the issues of supply chain, market systems, etc. makes the meaning of product family both broad and more complicated. In this paper we call such situation ‘global product family,’ and first characterizes its components and complexity. Following this, we proposes a mathematical model for the simultaneous design problem of module commonalization strategies under the given product architecture and supply chain configuration through selection of manufacturing sites for module production, assembly and final distribution as an instance of the problems. In the model, the choice of modules and various sites are represented with 0-1 design variables with the volume of production and transportation represented with non-negative continuous design variables, and the objective defined on total cost. An optimization method is configured with a genetic algorithm and a simplex method for such a mixed integer programming problem. Some numerical case studies are included to determine the validity and promise of the developed mathematical model and algorithm. Finally, we conclude with some discussion of future work.     

An integrated model for closed-loop supply chain configuration and supplier selection: Multi-objective approach

Reverse logistics consists of all operations related to the reuse of products. External suppliers are one of the important members of reverse logistics and closed loop supply chain (CLSC) networks. However in CLSC network configuration models, suppliers are assessed based on purchasing cost and other factors such as on-time delivery are ignored. In this research, a general closed loop supply chain network is examined that includes manufacturer, disassembly, refurbishing, and disposal sites. Meanwhile, it is managed by the manufacturer. We propose an integrated model which has two phases. In the first phase, a framework for supplier selection criteria in RL is proposed. Besides, a fuzzy method is designed to evaluate suppliers based on qualitative criteria. The output of this stage is the weight of each supplier according to each part. In the second phase, we propose a multi objective mixed-integer linear programming model to determine which suppliers and refurbishing sites should be selected (strategic decisions), and find out the optimal number of parts and products in CLSC network (tactical decisions). The objective functions maximize profit and weights of suppliers, and one of them minimizes defect rates. To our knowledge, this model is the first effort to consider supplier selection, order allocation, and CLSC network configuration, simultaneously. The mathematical programming model is validated through numerical analysis.     

Supply Chain Network Designs Developed for Deteriorating Items Under Conditions of Trade Credit and Partial Backordering

Deteriorating items, trade credit, and partial backordering are common in today’s business. However, no previous study on supply chain network design has considered these business aspects together. In this paper, we present supply chain networks designed for deteriorating items under trade credit and two conditions: (a) no shortage and (b) partial backordering of goods. We also present 2 algorithms based on nonlinear optimization that were developed in order to optimize the influence area and the joint replenishment-cycle time in the no-shortage case, and to identify the optimal shortage level in the partial-backordering case. The numerical examples presented herein illustrate how the solution procedure works. The effects of various values of the tested parameters on decisions and costs are also discussed. Our results could be used as a reference by managers when making business decisions.     

Close-optimal production and procurement policy for a X-network of added value using lexicographic linear goal programming

One of the main challenges of operation managers of firms is to setup feasible production and procurement plans. This is also the case of more complex structures such as supply chains. In almost all firms, specific tools like ERPs are used to support managers in their decision-making tasks. These tools manipulate huge amount of data such as the order backlog, or technical, marketing, suppliers and customers data. They work often based on the MRP principles and suggest production and procurement plans after a sequential procedure which begins by the material requirements calculation followed by the load balancing process. Very often, the load balancing is under the control of managers who try to take account of implicit constraints that cannot be modelled easily. This is a difficult and often risky task because the managers do not know what the best solution of the planning and procurement problem is. In other words, there is a lack of a kind of “the best feasible production and procurement target”. The main idea of this article is to suggest a complementary method for planning based on a specific mathematical programming approach which provides plans considering simultaneously all material and capacity constraints over the entire planning horizon. These plans can be considered as that necessary closed optimal production and procurement target for a company or a supply chain which uses an MRP-based planning tool. The lexicographic linear goal programming provides a suitable multi-criteria modelling paradigm for the production and procurement planning problems, especially for the supply chains. The study is focused on a common supply chain structure formed by several suppliers on one side and several customers on the other, connected together by a business-to-business relationship over a rolling horizon. The structure is modelled thanks to the Petri Nets supporting the definition of the global problem model. The model is then applied to a study case extracting from the car assembly industry.     

A Multiobjective Evolutionary Programming Algorithm and Its Applications to Power Generation Expansion Planning

The generation expansion planning (GEP) problem is defined as the problem of determining WHAT, WHEN, and WHERE new generation units should be installed over a planning horizon to satisfy the expected energy demand. This paper presents a framework to determine the number of new generating units (e.g., conventional steam units, coal units, combined cycle modules, nuclear plants, gas turbines, wind farms, and geothermal and hydro units), power generation capacity for those units, number of new circuits on the network, the voltage phase angle at each node, and the amount of required imported fuel for a single-period generation expansion plan. The resulting mathematical program is a mixed-integer bilinear multiobjective GEP model. The proposed framework includes a multiobjective evolutionary programming algorithm to obtain an approximation of the Pareto front for the multiobjective optimization problem and analytical hierarchy process to select the best alternative. A Mexican power system case study is utilized to illustrate the proposed framework. Results show coherent decisions given the objectives and scenarios considered. Some sensitivity analysis is presented when considering different fuel price scenarios.      

Two-echelon, multi-commodity supply chain network design with mode selection, lead-times and inventory costs

Designing distribution networks - as one of the most important strategic issues in supply chain management - has become the focus of research attention in recent years. This paper deals with a two-echelon supply chain network design problem in deterministic, single-period, multi-commodity contexts. The problem involves both strategic and tactical levels of supply chain planning including locating and sizing manufacturing plants and distribution warehouses, assigning the retailers' demands to the warehouses, and the warehouses to the plants, as well as selecting transportation modes.We have formulated the problem as a mixed integer programming model, which integrates the above mentioned decisions and intends to minimize total costs of the network including transportation, lead-times, and inventory holding costs for products, as well as opening and operating costs for facilities. Moreover, we have developed an efficient Lagrangian based heuristic solution algorithm for solving the real-sized problems in reasonable computational time.     

An agent-based distributed computational experiment framework for virtual supply chain network development

This paper focuses on research on virtual supply chain networks instead of real supply chain networks by making use of agent technology and computational experiment method. However, the recent research is inefficient in computational experiment modeling and lack of a related methodological framework. This paper proposes an agent-based distributed computational experiment framework with in-depth study of material flow, information flow and time flow modeling in supply chain networks. In this framework, a matrix-based formal representation method for material flow, a task-centered representation method for information flow and an agent-based time synchronization mechanism for time flow are proposed to aid building a high quality computational experiment model for a multi-layer supply chain network. In order to conduct the model, a computational experiment architecture for virtual supply chain networks is proposed. In this architecture, coordination mechanisms among agents based on material flow, information flow and time flow as well as consistency check methods for computational experiment models are discussed. Finally, an implementation architecture of the framework is given and a case of virtual supply chain network is developed to illustrate the application of the framework. The computational experiment results of the case show that the proposed framework, not only feasible but correct, has sound advantages in virtual supply chain network development, computational experiment modeling and implementation.     

Hierarchical Capacity Planning With Reconfigurable Kits in Global Semiconductor Assembly and Test Manufacturing

Kits (such as accessories, fixtures, jigs, etc.) are widely used in production for many industries. They are normally product- and machine-specific, so a large kit inventory must be maintained when the product-mix variation is high. Fortunately, many kits are reconfigurable. That means they can be dissembled into components and then these components themselves (or together with some other components) can be reassembled into new types of kits. Therefore, we can save money and improve supply chain responsiveness by purchasing components instead of entire kits. However, research on capacity planning with reconfigurable kits has not been reported. We proposed a two-level hierarchical planning methodology to generate a complete capacity planning solution using mixed-integer linear programming. MaxIt covers mid-range monthly planning and automated capacity allocation system covers short-range weekly planning. These systems are integrated to generate optimal capacity plans considering kit components. This methodology has been successfully implemented in Intel's global semiconductor assembly and test manufacturing since 2004. In this paper, we present the hierarchical modeling framework and focus on MaxIt modeling with kit reconfiguration. We also verify the methodology by numerical experiments in a real production environment.     

A multi-objective meta-heuristic approach for the design and planning of green supply chains - MBSA

Supply Chains are complex networks that demand for decision supporting tools that can help the involved decision making process. Following this need the present paper studies the supply chain design and planning problem and proposes an optimization model to support the associated decisions. The proposed model is a Mixed Integer Linear Multi-objective Programming model, which is solved through a Simulated Annealing based multi-objective meta-heuristics algorithm – MBSA. The proposed algorithm defines the location and capacities of the supply chain entities (factories, warehouses and distribution centers) chooses the technologies to be installed in each production facility and defines the inventory profiles and material flows during the planning time horizon. Profit maximization and environmental impacts minimization are considered. The algorithm, MBSA, explores the feasible solution space using a new Local Search strategy with a Multi-Start mechanism. The performance of the proposed methodology is compared with an exact approach supported by a Pareto Frontier and as main conclusions it can be stated that the proposed algorithm proves to be very efficient when solving this type of complex problems. Several Key Performance Indicators are developed to validate the algorithm robustiveness and, in addition, the proposed approach is validated through the solution of several instances.     

Model selection with considering the CO2 emission alone the global supply chain

This study formulates a model for analyzing eco-environmental impact on global supply chain network. The multi-criteria optimization model is applied to seek optimal solutions that not only can achieve predetermined objectives, but also can satisfy constraints for multi-product problems. The overall optimization is achieved using mathematical programming for modeling the supply chain functions such as location, inventory, production, distribution functions and transportation mode selections. Then, the supply chain model is formulated as a minimization problem for costs and environmental impacts. Herein, the solution is the flow of goods in global supply chain environment in different periods of time over one year. Furthermore, the numerical values obtained from a real company are applied to these mathematical formulations to test its usability. The testing is conducted in four different cases that include two combinations, no due date constraint and due date constraint, without connection of distributor and with connection of distributors. The results from these experiments can help in determining the best transportation routes, inventory levels, shipment quantity, and transportation modes. Specifically, the results propose a new configuration for designing global supply chain for the case company that could minimize economical and environmental impacts problems simultaneously.