Evaluating Sensor Placement Via Machine Learning Models
Abstract
As more sophisticated approaches to analyzing datasets become available the importance of sensor placement for expensive sensors increases. Datasets generated by these sensors help engineering managers take appropriate action in the early phases of the decision-making process but must be placed where the data provides the most value. among these high-cost sensors, water level gauge sensors are some of the most significant ones when it comes to flood hazard predictions, usually captured at stream gauge stations. the cost of installing these sensors has decreased over the last few decades, but they are still expensive enough to limit the number that can be placed. Pinpointing the best locations to place new sensors is still a challenge, as there are many other datasets that could influence the value of stream gauge information at a particular station. This study uses unsupervised learning techniques to evaluate sensor locations based on historical performance. the performance evaluation is conducted using unsupervised machine learning (divisive clustering) and supervised machine learning (deep-learning) models. the information gathered from the results of this research can help future research to pinpoint the appropriate locations for new sensor placement to improve the overall predictive capabilities of deep-learning models.
Recommended Citation
F. Sharafkhani and S. Corns, "Evaluating Sensor Placement Via Machine Learning Models," Proceedings of the American Society for Engineering Management 2023 International Annual Conference and 44th Annual Meeting: Climbing Higher with Engineering Management, pp. 646 - 653, American Society of Engineering Management, Jan 2023.
Department(s)
Engineering Management and Systems Engineering
Keywords and Phrases
deep learning; Sensor placement; unsupervised machine learning; water level gauge sensors
Document Type
Article - Conference proceedings
Document Version
Citation
File Type
text
Language(s)
English
Rights
© 2025 American Society of Engineering Management, All rights reserved.
Publication Date
01 Jan 2023
