Operational Analysis Revisited: Error Measure Limits of Assumptions

The assumptions used to develop operational analysiscomputer performance measures, such as number of jobs at adevice or response times, are stated in terms of the data itself,rather than the underlying system which produces the data. Inspite of claims of validity and as an aid in introducing queueingtheory in teaching, little has been written about operationalanalysis in the past ten years. Accuracy of operational analysisperformance measures depend on data behavior assumptionswhich can be validated with data based error measures.Increased soundness of the operational analysis approach may beobtained by determining the limits of assumption errors as thetime period of observation increases. Part I of this paper is areview of operational analysis and addresses some of the previousconcerns with its approach. Part II develops furtherunderstanding of operational analysis assumption errors byexamining their limits. Limits are found for the assumptionerrors of job flow balance, homogeneous arrivals andhomogenous services. While the job flow balance assumptionerror measure is shown to approach zero over time, thehomogeneity assumption error measures, in general, do not