Tractable (Q,R) heuristic models for constrained service levels

Article Abstract:

Tractable (Q,R) heuristic models have been developed for controlling finished goods inventory under continuous review when the service level is explicitly constrained. Q represents the order quantity, while R represents the reorder point. Two tractable closed-form heuristics are used, one based on the limiting behavior as the economic order quantity (EOQ) approaches + infinity, and another a power approximation based on a normal distribution for leadtime demand. The first performs better than using the traditional EOQ , while the second provides excellent cost performance. Other heuristics are shown to perform worse, especially when the EOQ is small relative to the standard derivation of leadtime demand.

author: Robinson, Lawrence W., Platt, David E., Freund, Robert B.
Inventory control

User Contributions:

Comment about this article or add new information about this topic:

CAPTCHA

Estimating market model betas: a comparison of random coefficient methods and their ability to correctly identify random variation

Article Abstract:

An evaluation of market models that employ random coefficient methods suggests that these methods cannot distinguish between significance and insignificance with respect to market model betas. It appears that the maximum likelihood method of market modeling has not been adequately tested, and that these models cannot identify market occurrences as being random coefficient processes. . This conclusion is supported by empirical testing and simulation.

author: McDonald, Bill
Measurement, Risk (Economics)

User Contributions:

Comment about this article or add new information about this topic:

CAPTCHA

Informational dynamics of censored observations

Article Abstract:

The characterization of families of distributions, which can keep tractable the analysis of stochastic models that involve censored observations of random variables, is derived. Both cumulative and discontinuous distributions are considered. The families of distributions presented are intended to guide modelers in the selection of distributions used and can assist in the modeling of informational dynamics driven by censored observations.

author: Braden, David J., Freimer, Marshall
Research, Distributions, Theory of (Functional analysis), Theory of distributions, Mathematical models

User Contributions:

Comment about this article or add new information about this topic:

CAPTCHA


subjects list: Models, Stochastic processes
This website is not affiliated with document authors or copyright owners. This page is provided for informational purposes only. Unintentional errors are possible.