Crowd sensitive indicators for proactive safety management: A theoretical framework

The Functional Resonance Analysis Method (FRAM) is a method to develop work domain models able of capturing the nitty-gritty of work. The analysis of different model instantiations supports the identification of possible misalignments between Work-As-Imagined (WAI) and Work-As-Done (WAD). Knowledge elicitation is currently a tricky stage for the modelling process, often being significantly time-consuming. A variety of knowledge elicitation techniques is thus used to compensate such information gap, including document analysis (procedures, standards, manuals), interviews (often open-ended or semi-structured), focus groups, workshops, and observations. Especially for the WAD, data gathering through interviews or mainly naturalistic observations becomes particularly time-consuming and labour-intensive since it has to reconcile multiple data sources (multiple operators performing tasks). For such purpose, we propose an IT framework designed to facilitate sharp-end operators’ WAD data gathering through a user-friendly app. Based on a FRAM-based ontology and knowledge elicited, a semantic reasoner named Creativity Machine will complement the data collection. The purpose of the framework consists of defining context-specific potential functional patterns (taking advantage of the FRAM structure). Such patterns will then be proposed to operators in order to receive confirmation about their significance in actual operating contexts. From these informative chunks, the framework will allow inferring multiple WAD instances, and associated context-specific leading indicators, i.e. H(CS)2 Is (Human-Centred Safety Crowd-Sensitive Indicators).