Maintenance indicators

Maintenance is a strategic process for many companies, supporting competitiveness and helping to control risks. It is no longer seen as a source of costs, but as a decisive means of increasing performance and profits. However, this goal can only be achieved if interventions are carried out according to the rules of the trade, in the right place, at the right time, and at a controlled overall cost.

The maintenance process must therefore be carefully controlled to remain as close as possible at all times to the optimum, which evolves under the influence of multiple external factors (variations: in demand and therefore in production, in labor costs, in the cost of raw materials and energy, etc.). The more efficient it is to maintain the production tool in the required state with a minimum of expenditure, the more competitive and profitable the company.

To monitor and achieve the objectives set, management needs to be able to rely on processes that adapt and improve continuously. For this reason, maintenance managers play an important role in the steering team, and they need measuring instruments. It's hard to imagine a factory control room or an aircraft cockpit without dynamic synoptics, dashboards, alarms and recorders for observing system behavior and the effects of actions taken. We therefore need to "instrument", i.e. collect useful information, express it in the form of indicators and, to better understand and interpret what they tell us, bring them together to create dashboards. In this way, maintenance can be managed with precision and efficiency, rather than navigating by sight with poor visibility.

Calculating indicators: yes, but which ones? And how do you interpret them, and then use them to take action? These are the questions we're going to try to answer. To do so, we'll start by positioning them within the management process, to distinguish the different stages that need to be completed to effectively improve performance. To clarify the multitude of indicators that can be encountered or defined, it will be useful to propose a classification according to the characteristics measured, and to specify their properties.

We then propose to break down maintenance into processes, so as to monitor the quality of their execution by determining the expressions of precise, targeted indicators. These indicators, once calculated, should be compared with reference values to assess maintenance activities. To establish a good diagnosis, it is also necessary to take into account contextual elements and data on equipment and plant malfunctions. It will then be possible to determine the improvement actions that need to be implemented to achieve the set objectives.