SIMISS: A Model-Based Searching Strategy for Inventory Management Systems

Inventory management is critical in human space flight operations. Currently, we use the inventory management system (IMS) in keeping track of items on the International Space Station (ISS). One challenge is to discover lost or wrongly placed items when IMS fails to discover them due to human factors. In this paper, we will illustrate a model-based searching strategy called semantic inventory management for ISS (SIMISS), with which possible locations of lost items will be calculated based on contextual features in three dimensions: (1) spatial; (2) temporal; and (3) human. It contains ontologies, databases, machine learning algorithms, and ubiquitous client applications. We have implemented and tested SIMISS with the sample data from IMS, operation data files and onboard short term plan experiments have been carried out in a set of simulation scenarios.