A Pitfall of estimated software cost
Masood Uzzafer
Abstract
Masood Uzzafer
Abstract
A pitfall hidden in the software cost estimation process is presented. Software project cost estimation process based on the quantile estimation software project cost distribution is not sub-additive and produce unrealistic cost estimates. Sub-additive is the process where the estimation of parts is less than or equal to the estimation of the whole. Further a solution is presented and illustrated how sub-additivity could be achieved by dealing with the expectation of the probability distribution beyond a given quantile.
OpenAlex reports 2 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
A pitfall hidden in the software cost estimation process is presented. Software project cost estimation process based on the quantile estimation software project cost distribution is not sub-additive and produce unrealistic cost estimates. Sub-additive is the process where the estimation of parts is less than or equal to the estimation of the whole. Further a solution is presented and illustrated how sub-additivity could be achieved by dealing with the expectation of the probability distribution beyond a given quantile.
Key concepts: Quantile, Computer science, Software, Cost estimate, Estimation, Software metric, Software quality, Process (computing)