The Internet of Things and Information Fusion: Who Talks to Who?
33 Pages Posted: 4 Mar 2018 Last revised: 26 Nov 2019
Date Written: February 1, 2018
Problem definition: The operational benefits of the Internet of Things (IoT) are predicated on the concept of autonomous sensors deployed by different firms providing real-time knowledge of the state of things. Sensors can improve their estimates by soliciting estimates from other sensors. The choice of which sensors to communicate with (``target'') is challenging because sensors (a) are constrained in the number of sensors they can target, and (b) only have partial knowledge of how other sensors operate (e.g., they do not know others' underlying inference algorithms/models). We examine the evolution of inter-firm sensor communication networks, investigate what patterns may emerge, and explore what drives such patterns.
Academic/Practical Relevance: The IoT will have a major disruptive impact on operations management (OM), and OM scholars need to develop and examine models and frameworks for analyzing sensor interactions. This is especially important because sensor communication builds ties across firms that require ongoing management.
Methodology: Analytic modeling combining estimation, optimization, and dynamic learning.
Results: We show that when selecting its target(s) each sensor needs to consider both the quality of the other sensors and its level of trust (i.e, understanding) in their inference models. Importantly, we establish that the state of the environment plays a key role in mediating quality and trust. When sensor qualities are public, we show that each sensor eventually settles on a constant target set but this long run target set is sample-path dependent and varies by sensor. The long run network, however, can be fully defined at time zero as a random directed graph, and hence one can probabilistically predict it. This prediction can be made perfect (i.e., the network can be identified in a deterministic way) after observing the state values for a limited number of periods. When sensor qualities are private, our results reveal that sensors may not settle on a constant target set.
Managerial Implications: Our work allows managers to not only predict which other firms their sensors will interact with but to also influence the outcomes through levers such as sensor quality and initial trust. This predictive ability and managerial control is important given that sensor communications build organizational ties that require attention and resources.
Keywords: Robust Estimation, Data Fusion, Information Sharing
Suggested Citation: Suggested Citation