• Written by Sonali Subhadarsini, Product Manager

A typical Monday morning for a maintenance team in energy and utilities often begins with a flurry of activity, as they scramble to respond to unexpected equipment failures or prepare for scheduled downtime. The shift supervisor checks the maintenance log, while the maintenance planner reviews the work order schedule, and the reliability engineer analyzes condition data from critical assets. But amidst all this activity, there are often questions that go unanswered, or at least, unanswerable in a timely manner, because they depend on a report that is still being compiled.

The maintenance questions that currently wait on a report

These questions can range from simple queries, such as which assets are due for maintenance, to more complex ones, like which spare parts are currently in stock, or what is the average mean time between failures for a particular type of equipment. The maintenance team may also want to know which work orders are pending, or which assets have been taken offline for maintenance. Currently, these questions often have to wait until a report is generated, which can take hours, or even days, depending on the complexity of the query and the availability of the data.

The category manager, responsible for procuring spare parts, may need to know which suppliers have the required parts in stock, and at what price. The maintenance planner, on the other hand, may want to know which maintenance activities can be scheduled during a planned outage, and which ones need to be deferred. These questions are critical to the smooth operation of the maintenance team, but they often have to wait until the report is ready.

The data behind them: work orders, asset history, spare parts inventory, condition data where it exists

The data required to answer these questions is often scattered across different systems, including the computerized maintenance management system (CMMS), the enterprise resource planning (ERP) system, and the condition monitoring system. The CMMS contains information on work orders, asset history, and maintenance schedules, while the ERP system has data on spare parts inventory, procurement, and supplier information. The condition monitoring system, where it exists, provides real-time data on the condition of critical assets.

Integrating this data can be a challenge, especially when the systems are not designed to talk to each other. The maintenance team may have to rely on manual data entry, or use cumbersome data extraction tools to get the information they need. This can lead to delays, and even errors, which can have serious consequences in a sector where safety and reliability are paramount.

Sector constraints: outage planning and regulatory reporting burden

The energy and utilities sector is heavily regulated, and maintenance teams have to contend with a complex web of rules and regulations. Outage planning, for example, requires careful coordination with regulatory bodies, as well as with other stakeholders, such as grid operators and customers. The maintenance team has to ensure that outages are planned and executed in a way that minimizes disruption to the grid, and that all regulatory requirements are met.

Regulatory reporting is another significant burden on maintenance teams. They have to generate reports on maintenance activities, asset performance, and compliance with regulatory requirements. These reports can be time-consuming to generate, and may require significant manual effort to compile the required data. The production of these reports can take away from the time available for actual maintenance activities, which can have serious consequences for asset reliability and safety.

What improvement looks like: planned versus reactive ratio, mean time between failures, spares availability

Improving maintenance operations in energy and utilities requires a focus on key performance indicators, such as the planned versus reactive ratio, mean time between failures, and spares availability. The planned versus reactive ratio measures the proportion of maintenance activities that are planned, versus those that are reactive. A high planned versus reactive ratio indicates that the maintenance team is able to anticipate and prevent equipment failures, rather than simply responding to them.

The mean time between failures is another important metric, as it measures the average time between equipment failures. A longer mean time between failures indicates that the maintenance team is able to maintain assets in a way that prevents frequent failures. Spares availability is also critical, as it ensures that the maintenance team has the required parts to perform maintenance activities in a timely manner. By focusing on these metrics, maintenance teams can improve asset reliability, reduce downtime, and minimize the risk of regulatory non-compliance.

What this does not do, or when it is the wrong choice

While AI for maintenance in energy and utilities can be a powerful tool, it is not a panacea for all maintenance ills. It is not a replacement for skilled maintenance personnel, nor is it a substitute for proper maintenance procedures. It is also not a solution for underlying problems, such as inadequate training, or poor asset design. In these cases, the contact us to discuss how AsscherAi can help with specific maintenance challenges, and to determine whether AI is the right solution for your organization.

Most people get wrong about AI for maintenance that it can somehow magically fix all maintenance problems, without any effort or investment from the maintenance team. This is wrong, as AI is simply a tool that can help maintenance teams make better decisions, and work more efficiently. It requires careful planning, implementation, and ongoing maintenance to ensure that it is used effectively.