Premium
Interval reliability of spare part stocks
Author(s) -
Bazovsky Igor,
Benz Glen
Publication year - 1988
Publication title -
quality and reliability engineering international
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.913
H-Index - 62
eISSN - 1099-1638
pISSN - 0748-8017
DOI - 10.1002/qre.4680040307
Subject(s) - spare part , poisson distribution , reliability engineering , reliability (semiconductor) , interval (graph theory) , computer science , limiting , pipeline (software) , stockout , operations research , engineering , statistics , operations management , mathematics , mechanical engineering , power (physics) , physics , quantum mechanics , combinatorics , programming language
Probability goals are commonly used in conjunction with Poisson‐distributed variability to estimate initial pipeline spares requirements for aircraft organizations. An analysis of field data indicates that these deployment models may significantly underestimate spares acquisition quantities required in order to maintain probability goals throughout a system life cycle. This paper defines reliability of spare part stocks as the probability that these stocks will satisfy demands throughout a calendar time interval which begins at random points in the life cycle. Five models for spares planning were applied to field data from a small aircraft organization. The common goal in these analyses was based on military logistic standards: 0.99 probability for each part type over a replenishment interval of 360 h. Two initial interval (Poisson) models produced spares requirements averaging 273 units. The average for the three limiting (non‐Poisson) models was 3630 units. The limiting models also made it evident that spares must be purchased to refill the pipeline several times in a 20 year life cycle. It is concluded that a limiting buffer stock model is most realistic for planning normal spares requirements and that a renewal model should be used if the pipeline could be severed during surge periods of activity.