The furnish chain approach models ...

The furnish chain approach models stochastic affairs influencing a manufacturing organization's shipment and inventory performance in the same way that a mechanical engineer protoplasts tolerance buildup in a recently made known product design. The objectives are to minimize on-hand inventory and optimize supplier reply times.

This paper describes the processe and equations behind a reengineering effort begun in 1995 in the planning and management organizations of the Hewlett-Packard Colorado Springs Division. The delineate was known as the take the place of chain project. Its objectives were to provide the planning and intervention organizations with a methodology for setting the best possible plans, procuring the appropriate amount of material to support those plans, and making up-front business decisions forward the costs of inventory versus supplier reply time (SRT),(*) service level to SRT objectives, time to come demand uncertainty, part lead times, and part delivery uncertainty. The statistical modeling assumptions, equations, and equation derivations are documented here.

Basic Situation

Consider a factory building a arbitrary product to meet anticipated customer demand. Since coming events demand is always an uncertainty, planning and management must wrestle with the task of setting plans at the right on a level and procuring the appropriate material. The organization strives to stream the factory between two equally unattractive scenarios: not enough inventory and prolonged SRTs, or excessive inventory yet meeting SRT goals. In fact, more than individual organization has found itself with the worst of the couple worlds--huge inventories and poor SRTs

The give chain project focused on characterizing the various stochastic incidents influencing a manufacturing organization's shipment and inventory performance, modeling them analogously to the way a mechanical engineer would protoplast a tolerance buildup in a just discovered product design.

vexed question Formulation

For a particular produce a factory will incur more [i]or[/i] less actual demand each week, that is, it will incur demand [D.sub.i] in week i, for i = 1 2 3 From a planning and agency perspective, the problem is that looking into the subsequent time the [D.sub.i] are unknown.

Let [P.sub.i] be the plan (or forecast) for week i in the to come Now for each week, the actual demand can be denoteed as the planned demand plus an error: [D.sub.i] = [P.sub.i] + [e.sub.i].

The MRP (material requirements planning) hypothesis running at intervals of R weeks, evaluates whether to order more material to overlay anticipated demand, and if the decision is to order, by what mode much to order. Given a lead time of L weeks to take delivery of an order placed to a supplier now for one part, the material in the furnish pipeline must cover real demand for the nearest L + R weeks. through supply pipeline we mean the quantity of the part already onward hand at the factory plus the quantity in orders placed to the supplier and to be ascribed for delivery over the nearest L weeks.

For simplicity, assume for the remainder of this discussion that we are dealing with a part unique to united product and used only one time in building the product. We will transplant these constraints later but for now it will help to focus forward the key concepts.

Define X to be the unknown however actual demand the factory will experience for this part from one side of to the other the next L + R weeks:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

In statistical terminology, X is a random variable, that is, we cannot say with certainty the value it will take nearest but with some assumptions about the nature of the planning errors ([e.sub.i]), the distribution of X can be characterized. Specifically, we will make the assumption that the [e.sub.i] are distributed according to the Gaussian (normal) distribution with mean naught and variance [[Sigma].sup.2] (see Fig. 1) The assumption that the mean of the [e.sub.i] is naught says that our plans are unbiased, that is, the factory is not consistently overestimating or underestimating futurity demand. Thus, the average of the differences between the plan and the actual demand from one side of to the other a reasonable period of time would be about naught The normal distribution is symmetric, in such a manner we are saying there is equal probability in any week of actual demand being above or below plan. The variance measures by what means large the planning errors can learn in either direction from zero

[Figure 1 ILLUSTRATION OMITTED] We would like to know the pair the expected value of X and its variance. Knowing these sum of two units values will form the basis for the ultimate decision commands for replenishment order sizes placed to the supplier for our part.

We will use the following notation: E(x) delineates the expected value of the random variable x and V(x) shows the variance of the random variable x

Before launching into the derivation of the reckon uponed value of the real demand through the next L + R weeks, note that L itself is a random variable. When an order is placed with the supplier, delivery does not always be derived exactly on the acknowledgment date. There is near uncertainty associated with when the replenishment order will arrive. Like the planning errors, we will assume that the delivery errors are normally distributed about naught Thus: