A Markov Model of the Learning Curve

Harry Larsen

MPRA Paper from University Library of Munich, Germany

Abstract: A learning curve is the labor hours per unit of production in the production sequence. Before its cost realization, it is typically forecast using the power function, hours(unit) = a unitb, with a and b derived from historical data via regressions. After the first unit’s cost is realized, future costs can be projected from the sequence of actualized units. This paper transforms the power function into a Markov model and applies a Kalman Filter to estimate the expected value, slope, and variance of a learning curve projection. To exemplify the solution, the parameters of the Kalman Filter are estimated from a set of twenty 1000-unit learning curves derived from The Theory of Complex Work. For those responsible for predicting future values of an ongoing production program, this procedure provides optimal estimates of unit labor hours and their variances.

Keywords: Markov; Kalman Filter; Learning Curve; Excel; Solver (search for similar items in EconPapers)
JEL-codes: C2 C5 C53 L6 L60 M2 M21 (search for similar items in EconPapers)
Date: 2026-03-22
References: Add references at CitEc
Citations:

Published in Journal of Cost Analysis & Parametrics 1.13(2026): pp. 104-113

Downloads: (external link)
https://mpra.ub.uni-muenchen.de/128435/1/MPRA_paper_128435.pdf original version (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:pra:mprapa:128435

Access Statistics for this paper

More papers in MPRA Paper from University Library of Munich, Germany Ludwigstraße 33, D-80539 Munich, Germany. Contact information at EDIRC.
Bibliographic data for series maintained by Joachim Winter ().