Equipment criticality assessment — Metrics and performance indicators

This article presents the metrics and performance indicators used in different industrial sectors to assess and/or help reduce the criticality of the consequences of equipment failures.

It is essential that company managers and the bodies responsible for establishing and enforcing regulations have dashboards to monitor the effectiveness of the measures put in place to prevent or minimize the consequences of equipment failure.

The term metric used in this article comes from the Anglicism of the word "metrics" and refers to measurement methods by which the effectiveness of a process or product can be assessed.

The vast majority of these monitoring tools have been defined to quantify and qualify the impact of critical failures on the economic profitability of investments, the safety of goods and people, operational availability, compliance with regulatory constraints for hazardous installations and the environment. Given the proliferation of metrics and performance indicators, it is always difficult for the uninformed user to select the most relevant set of indicators, and this article is intended as a guide.

An in-depth analysis of their definitions enables us to divide their fields of application into two categories: on the one hand, the field of traditional, non-hazardous activities for the legislator, and on the other, the field of hazardous, classified activities.

Two families of performance indicators have been defined for each of these categories.

The first family is designed to evaluate performance calculated from data collected on installations: it provides information on the consequences of upstream management decisions. These performance indicators are called "lagging indicators" in the English-speaking world, and are referred to in this article as "impact indicators".

The second family concerns indicators linked to technical or organizational measures that ultimately help to reduce the criticality of a failure. Referred to as "leading indicators", they will hereafter be called "activity indicators".

The notion of equipment criticality has many, often ambiguous, interpretations; the first part of this section will propose a definition. As failures do not necessarily have the same consequences depending on the sector of activity, a classification of conventional sectors of activity and sectors of activity classified as hazardous will be presented, together with the associated regulations for the latter category. Next, a typology of failures will identify three classes of failure: organizational, human and technical. As organizational and human failures predominate, James Reason's "Swiss cheese" model will be...