摘要:

AbstractWe consider a model in which when a device fails it is either repaired to its condition prior to failure or replaced. Moreover, the device is replaced at timeskT, k= 1, 2, … The decision to repair or replace the device at failure depends on the age of the device at failure. We find the optimal block time,T0, that minimizes the long‐run average cost of maintenance per unit time. Our results are shown to extend many of the well known results for block replacement polic

ISSN:8755-0024    

DOI:10.1002/asm.3150030203

出版商:John Wiley&Sons, Ltd.    

年代:1987

数据来源: WILEY

摘要:

AbstractIn their article, entitled ‘Group technology in production management, the short horizon planning level’, H. Garcia and J. M. Proth have stated the following problem: starting from a (0, 1) binary matrix of size (N x M), how to divide into independent subsets the rows of this matrix simultaneously with a one‐to‐one corresponding partition of the columns, maximizing the presence of 1s in the intersecting blocks with a joint minimization of the presence of 0s outside of these blocks.The authors have proposed an efficient and very fast heuristic algorithm in comparison with the existing methods of a fast‐growing literature on the subject. The only drawback of this algorithm is its dependence on the initial partition. In this paper, we try to improve this algorithm slightly, first in rewriting the objective function in a linear form and secondly in giving computational improvements related to this linear fo

ISSN:8755-0024    

DOI:10.1002/asm.3150030204

出版商:John Wiley&Sons, Ltd.    

年代:1987

数据来源: WILEY

摘要:

AbstractA theoretical model of interpurchase times at the individual level is developed. This model provides a bench‐mark for the expected regularity of a product's purchase cycle in the absence of any confounding external marketing activity. Products which empirically deviate from the theoretical expectations can then be further studied to determine the causes of the deviations. The theoretical model also tests the robustness of the assumption of a gamma distribution at the individual level, an assumption which is frequently made in the marketing literatur

ISSN:8755-0024    

DOI:10.1002/asm.3150030205

出版商:John Wiley&Sons, Ltd.    

年代:1987

数据来源: WILEY

摘要:

AbstractWe present a general framework for treating categorical data with errors of observation. We show how both latent class models and models for doubly sampled data can be treated as exponential family nonlinear models. These are extended generalized linear models with the link function substituted by an observationwise defined non‐linear function of the model parameters. The models are formulated in terms of structural probabilities and conditional error probabilities, thus allowing natural constraints when modelling errors of observation. We use an iteratively reweighted least squares procedure for obtaining maximum likelihood estimates. This is faster than the traditionally used EM algorithm and the computations can be made in GLIM.1As examples we analyse three sets of categorical data with errors of observation which have been analysed before by Ashford and Sowden,2Goodman3and Chen,4respectivel

ISSN:8755-0024    

DOI:10.1002/asm.3150030206

出版商:John Wiley&Sons, Ltd.    

年代:1987

数据来源: WILEY

ISSN:8755-0024    

DOI:10.1002/asm.3150030202

出版商:John Wiley&Sons, Ltd.    

年代:1987

数据来源: WILEY