Psychological artificial intelligence: Designing algorithms to deal with the uncertainty of rework in construction
As construction organizations are confronted with uncertainty and imperfect information, they find accommodating the likelihood of rework in their projects challenging. Bayesian statistical models cannot be utilized to predict rework as objective, and even subjective probabilities are unknown. In un...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2024-12-01
|
| Series: | Developments in the Built Environment |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2666165924002679 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | As construction organizations are confronted with uncertainty and imperfect information, they find accommodating the likelihood of rework in their projects challenging. Bayesian statistical models cannot be utilized to predict rework as objective, and even subjective probabilities are unknown. In uncertainty settings, algorithms such as smart heuristics – simple task-specific decision strategies that function under specific conditions – have been shown to achieve equal and better performance in problems of inference than machine learning models. However, algorithms to effectively deal with the uncertainty of rework in construction have yet to be developed. Hence, the motivation for this paper is to examine how psychological artificial intelligence, which applies insights from psychology (e.g., mental and social processes) to design algorithms, can be potentially used to develop smart heuristics that can cater to the uncertainty of rework in construction in varying conditions and contexts. To this end, the contributions of this paper are twofold as it: (1) brings to the fore a new line of inquiry to deal with not only the uncertainty of rework using psychological insights to design simple algorithms but also unexpected events in general; and (2) provides guidance to ensure the design of algorithms to deal with the uncertainty that reflects the actualities of practice. |
|---|---|
| ISSN: | 2666-1659 |