Interpreting Engineering Program Costs Using Explainable AI

DOI

10.2991/978-94-6463-978-0_42How to use a DOI?

Keywords

Explainable AI; SHAP; LIME; Cost forecasting; Engineering program management; Ensemble learning

Abstract

Accurate cost estimation in engineering projects is vital for effective planning and risk analysis. Traditional analytical models yield high precision but compromise on interpretability, hence limiting managerial confidence. In this paper, a reproducible, Python-based explainable-AI (XAI) workflow integrating tree-ensemble-based regressors like XGBoost and Random Forest with SHAP (Shapley Additive Explanations) and LIME (Local Interpretable Model-Agnostic Explanations) is presented to account for the drivers of costs in engineering projects. Two publicly available datasets - a dataset of software costs and a manufacturing defects dataset are subjected to cross-domain testability checks. Findings reveal interpretable cost estimation is achievable without sacrificing predictive accuracy (MAPE < 10 %) while having global and local interpretability of feature contributions. The model showcases how cost drivers such as function points, code length, defect rate, and maintenance hours account for overall costs, enabling data-driven managerial decisions.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Shreya Makinani
AU  - Pankaj Siri Bharath Bairu
PY  - 2025
DA  - 2025/12/31
TI  - Interpreting Engineering Program Costs Using Explainable AI
BT  - Proceedings of the 1st Engineering Data Analytics and Management Conference (EAMCON 2025)
PB  - Atlantis Press
SP  - 487
EP  - 499
SN  - 2352-5401
UR  - https://doi.org/10.2991/978-94-6463-978-0_42
DO  - 10.2991/978-94-6463-978-0_42
ID  - Makinani2025
ER  -