Learning automata decision analysis for sensor placement. Issue 9 (2nd September 2018)

• Record Type: Journal Article
• Authors: Ben-Zvi, Tal
• Is Part Of: Journal of the Operational Research Society. Volume 69:Issue 9(2018)
• Page Start: 1396
• Page End: 1405
• Publication Date: 2018-09-02
• Subjects: Decision analysis -- learning automata -- optimisation -- sensor placement; Operations research -- Periodicals; 658.4034
• DOI: 10.1080/01605682.2017.1398205
• Languages: English
• ISSNs: 0160-5682

Abstract

Abstract: This study investigates how to design sensor systems in a way that responds to certain factors in the environment. This decision analysis problem focuses on sensor placement: how to place sensors to find an intruder that is affected by environmental elements. The sensors we use are of two types: the first type detects targets, and the second type detects elements in the environment. Techniques from the learning automata literature are used to develop a detection mechanism. The approach proposed in this study is dynamic, and can adjust to environmental variations. And its rate of detection exceeds static approaches, such as evenly spread sensor configuration. This work has implications for the design of any sensor system in which the physical environment shapes the probability of events occurring.