Gerald E. Feigin

A Supply Network Model with Base-Stock Control and Service Requirements

We develop a supply network model that takes as input the bill of materials, the (nominal) lead times, the demand and cost data, and the required customer service levels. In return, the model generates the base-stock level at each store--the stocking location for a part or an end-product, so as to minimize the overall inventory capital throughout the network and to guarantee the customer service requirements. The key ingredient of the model is a detailed, albeit approximate, analysis of theactual lead times at each store and the associated demand over such lead times, along with a characterization of the operation at each store via an inventory-queue model. The gradients are derived in explicit forms, and a conjugate gradient routine is used to search for the optimal solution. Several numerical examples are presented to validate the model and to illustrate its various features.

A queueing network model for semiconductor manufacturing

We develop an open queueing network model for rapid performance analysis of semiconductor manufacturing facilities. While the use of queueing models for performance evaluation of manufacturing systems is not new, our approach differs from others in the detailed ways in which we model the different tool groups found in semiconductor wafer fabrication, as well as the way in which we characterize the effect of rework and scrap on wafer lot sizes. As an application of the model, we describe a method for performing tool planning for semiconductor lines. The method is based on a marginal allocation procedure which uses performance estimates from the queueing network model to determine the number of tools needed to achieve a target cycle time, with the objective being to minimize overall equipment cost.

Inventory Planning in Large Assembly Supply Chains

Large assembly supply chains, such as those found in the computer, consumer electronics, and automobile industries, usually support the production of multiple end-products where each end-product has a complex multi-level bill of material (BOM) consisting of hundreds, if not thousands, of components and subassemblies with widely varying lead times and costs. The end-products typically have many of these components and subassemblies in common. The supply chains are subject to demands for end-products which are highly volatile and notoriously difficult to forecast, yield and other quality problems, periodic engineering changes, frequent new product introductions, rapid obsolescence of end-products and components, and geographically dispersed production and vendor locations.

Scheduling Semiconductor Lines Using A Fluid Network Model

We present a novel meth.od f o r “what’s next” scheduhg of semiconductor manufacturing lines based on the determanistic fluid network model of Chen and Yao [l]. By “what’s next” scheduling we mean specifying the order i n which jobs are processed at each tool group. The approach we take here is first to determine how to allocate tool capacity among competing job types b y solving a series of linear and quadratic programming problems related to the fluid model and then, to specify a “what’s next” scheduling algorithm design,ed to track these capacity allocations. The primary advantage of our approach. is that it gives rise to a schedule which is based on global rath,er than local state informadion and which, is responsive to stochastic changes an the line including tool incapacitation events and operator unaraila,bility. In addition to describing the scheduling algorithm, we present some results about the fluid model that have important implications in the context of semiconductor manufacturing.

Capacity allocation in semiconductor fabrication

One of the most serious challenges in the semiconductor memory business is the rapid price decline. We develop an allocation scheme that determines the die (chip) allocation among different memory products. The allocation takes into account available die capacity, customer service requirements, as well as price declines and demand distributions among different products.

DISTRIBUTION RESOURCE PLANNING SYSTEMS: A CRITIQUE AND ENHANCEMENT

Distribution Resource Planning (DRP) is a general framework for planning and managing inventory in distribution networks. The DRP framework can be applied to complex distribution networks with thousands of unique stock-keeping units and hundreds of stocking locations. It allows for non-stationary (e.g. seasonal) demand patterns and a wide variety of user-specified inventory control rules including all standard inventory policies such as (S, s) and fixed order quantity rules. A number of software implementations of DRP are commercially available and are widely used in industry. In this paper, we describe the logic underlying DRP and point out some of its limitations. The inner workings of DRP are not always familiar to the research/academic community. On the other hand, practitioners may be unaware of some of the shortfalls and limitations of the system. Our objective here is to bridge this gap. In particular, we show how the performance evaluation capability of DRP can be substantially enhanced by some simple analytical formulas, derived as approximations from base-stock and (S, s) control schemes.

A framework for planning and scheduling of manufacturing enterprises

Most control models of manufacturing systems do not adequately address the complex problems that modern manufacturing enterprises face. Many models oversimplify manufacturing operations and objectives, ignore customer oriented objectives, emphasize optimal control at the expense of robust, realistic models of production, and fail to tie stated control objectives to overall business objectives. We describe a framework for control of manufacturing enterprises that encompasses a wider scope of activities than is typically considered in control models. Our framework is a loosely coupled hierarchy in which information is shared among different levels of the hierarchy and decisions at one level can serve as constraints at other levels. We outline the structure of this framework, specifying for each level the time horizon and the relevant decisions, objectives, and constraints.