Optimization-Based Available-To-Promise with Multi-Stage Resource Availability

Increasingly, customer service, rapid response to customer requirements, and flexibility to handle uncertainties in both demand and supply are becoming strategic differentiators in the marketplace. Organizations that want to achieve these benchmarks require sophisticated approaches to conduct order promising and fulfillment, especially in today’s high-mix low-volume production environment. Motivated by these challenges, the Available-to-Promise (ATP) function has migrated from a set of availability records in a Master Production Schedule (MPS) toward an advanced real-time decision support system to enhance decision responsiveness and quality in Assembly To Order (ATO) or Configuration To Order (CTO) environments. Advanced ATP models and systems must directly link customer orders with various forms of available resources, including both material and production capacity. In this paper, we describe a set of enhancements carried out to adapt previously published mixed-integer-programming (MIP) models to the specific requirements posed by an electronic product supply chain within Toshiba Corporation. This model can provide individual order delivery quantities and due dates, together with production schedules, for a batch of customer orders that arrive within a predefined batching interval. The model considers multi-resource availability including manufacturing orders, production capability and production capacity. In addition, the model also takes into account a variety of realistic order promising issues such as order splitting, model decomposition and resource expediting and de-expediting. We conclude this paper with comparison of our model execution results vs. actual historical performance of systems currently in place.     

A decision support system for order acceptance/rejection in hybrid MTS/MTO production systems

In this paper, we present a novel decision support system for order acceptance/rejection in a hybrid Make-to-Stock/Make-to-Order production environment. The proposed decision support system is comprised of five steps. At the first step, the customers are prioritized based on a fuzzy TOPSIS method. Rough-cut capacity and rough-cut inventory are calculated in the second step and in case of unavailability in capacity and materials, some undesirable orders are rejected. Also, proper decisions are made about non-rejected orders. At the next step, prices and delivery dates of the non-rejected orders are determined by running a mixed-integer mathematical programming model. At the fourth step, a set of guidelines are proposed to help the organization negotiate over price and due date with the customers. In the next step, if the customer accepts the offered price and delivery date, the order is accepted and later considered in the production schedule of the shop floor, otherwise the order is rejected. Finally, numerical experiments are conducted to show the tractability of the applied mathematical programming model.     

An integrated production and inventory model to dampen upstream demand variability in the supply chain

We consider a two-echelon supply chain: a single retailer holds a finished goods inventory to meet an i.i.d. customer demand, and a single manufacturer produces the retailer’s replenishment orders on a make-to-order basis. In this setting the retailer’s order decision has a direct impact on the manufacturer’s production. It is a well known phenomenon that inventory control policies at the retailer level often propagate customer demand variability towards the manufacturer, sometimes even in an amplified form (known as the bullwhip effect). The manufacturer, however, prefers to smooth production, and thus he prefers a smooth order pattern from the retailer. At first sight a decrease in order variability comes at the cost of an increased variance of the retailer’s inventory levels, inflating the retailer’s safety stock requirements. However, integrating the impact of the retailer’s order decision on the manufacturer’s production leads to new insights. A smooth order pattern generates shorter and less variable (production/replenishment) lead times, introducing a compensating effect on the retailer’s safety stock. We show that by including the impact of the order decision on lead times, the order pattern can be smoothed to a considerable extent without increasing stock levels. This leads to a situation where both parties are better off.     

Strategic capacity planning for light emitting diode (LED) supply chains across Taiwan and China

This paper describes a mixed-integer programming (MIP) model formulated for strategic capacity planning for light emitting diode (LED) makers of Taiwan, major companies in the global LED market. These firms have complex supply chains across Taiwan and China, and the region’s unique political and economic environment has created not only competitive advantages but also challenges in supply chain management: government regulations require that customer orders be accepted from Taiwan or China according to customer attributes; when conducting manufacturing, Taiwanese firms may need to transfer orders across national borders for reasons such as manufacturing technology (the required technology is available only at certain manufacturing facilities) or more efficient capacity utilization; and there are operations to be performed with specific processing requirements to follow, posing substantial challenges for planners. Motivated by the significance of these firms in the global market, we develop a MIP model with novel features to support their strategic capacity planning, covering demand and manufacturing-related decisions, including order acceptance and transfer, manufacturing starts, capacity expansion, and logistics. We illustrate the model’s performance using modified industry data in a numerical example; we also describe the potential impacts the model may create in industry applications.     

Supplier selection in make-to-order environment with risks

The problem of allocation of orders for parts among part suppliers in a customer driven supply chain with operational risk is formulated as a stochastic single- or bi-objective mixed integer program. Given a set of customer orders for products, the decision maker needs to decide from which supplier to purchase parts required for each customer order to minimize total cost and to mitigate the impact of delay risk. The selection of suppliers and the allocation of orders is based on price and quality of purchased parts and reliability of on time delivery. To control the risk of delayed supplies, the two popular percentile measures of risk are applied: value-at-risk and conditional value-at-risk. The proposed approach is capable of optimizing the supply portfolio by calculating value-at-risk of cost per part and minimizing mean worst-case cost per part simultaneously. Numerical examples are presented and some computational results are reported.     

Order planning for networked make-to-order enterprises—a case study

From the general trend towards global markets and a growing customer orientation, new concepts and forms of organisation are emerging, such as distributed or networked enterprises. One key requirement of these new paradigms is the availability of models and tools to support order negotiation with the optimisation of manufacturing routes and logistics and ensuring the co-ordination of all participating entities. We address the problem of planning an incoming customer order to be produced in a distributed (multi-site) and multi-stage production system. In particular, we have used as a case study the industry of semiconductors (in the business area of application specific integrated circuits). The problem is tackled in a hierarchical model, in two levels: there is a global network planning procedure, and a set of local capacity models associated to the different production units reflecting their particular features. An approach based on simulated annealing is presented, as well as a specially designed constructive heuristic, that takes into account many of the real world constraints and complexities. The general performance of the simulated annealing algorithm is assessed through some preliminary computational experiments. Finally, some concluding remarks and current directions of research are presented.     

Reinforcement learning versus heuristics for order acceptance on a single resource

Order Acceptance (OA) is one of the main functions in business control. Accepting an order when capacity is available could disable the system to accept more profitable orders in the future with opportunity losses as a consequence. Uncertain information is also an important issue here. We use Markov decision models and learning methods from Artificial Intelligence to find decision policies under uncertainty. Reinforcement Learning (RL) is quite a new approach in OA. It is shown here that RL works well compared with heuristics. It is demonstrated that employing an RL trained agent is a robust, flexible approach that in addition can be used to support the detection of good heuristics.     

Optimization of production allocation and transportation of customer orders for a leading forest products company

A leading manufacturer of forest products with several production facilities located in geographical proximity to each other has recently acquired a number of new production plants in other regions/countries to increase its production capacity and expand its national and international markets. With the addition of this new capacity, the company wanted to know how to best allocate customer orders to its various mills to minimize the total cost of production and transportation. We developed mixed-integer programming models to jointly optimize production allocation and transportation of customer orders on a weekly basis. The models were run with real order files and the test results indicated the potential for significant cost savings over the company’s current practices. The company further customized the models, integrated them into their IT system and implemented them successfully. Besides the actual cost savings for the company, the whole process from the initial step of analyzing the problem, to developing, testing, customizing, integrating and finally implementing the models provided enhanced intelligence to sales staff.     

Semiconductor lot allocation using robust optimization

In this work, the problem of allocating a set of production lots to satisfy customer orders is considered. This research is of relevance to lot-to-order matching problems in semiconductor supply chain settings. We consider that lot-splitting is not allowed during the allocation process due to standard practices. Furthermore, lot-sizes are regarded as uncertain planning data when making the allocation decisions due to potential yield loss. In order to minimize the total penalties of demand un-fulfillment and over-fulfillment, a robust mixed-integer optimization approach is adopted to model is proposed the problem of allocating a set of work-in-process lots to customer orders, where lot-sizes are modeled using ellipsoidal uncertainty sets. To solve the optimization problem efficiently we apply the techniques of branch-and-price and Benders decomposition. The advantages of our model are that it can represent uncertainty in a straightforward manner with little distributional assumptions, and it can produce solutions that effectively hedge against the uncertainty in the lot-sizes using very reasonable amounts of computational effort.     

A multi-stage supply chain network design problem with in-house production and partial product outsourcing

We propose a mixed-integer linear programming model for a novel multi-stage supply chain network design problem. Our model integrates location and capacity choices for plants and warehouses with supplier and transportation mode selection, and the distribution of multiple products through the network. The aim is to identify the network configuration with the least total cost subject to side constraints related to resource availability, technological conditions, and customer service level requirements. In addition to in-house manufacturing, end products may also be purchased from external sources and consolidated in warehouses. Therefore, our model identifies the best mix between in-house production and product outsourcing. To measure the impact of this strategy, we further present two additional formulations for alternative network design approaches that do not include partial product outsourcing. Several classes of valid inequalities tailored to the problems at hand are also proposed. We test our models on randomly generated instances and analyze the trade-offs achieved by integrating partial outsourcing into the design of a supply chain network against a pure in-house manufacturing strategy, and the extent to which it may not be economically attractive to provide full demand coverage.