The Importance of Timing and Data Accuracy for Asset Management

As we continue the journey into the digital age, data is becoming more important than ever. The latest trends in manufacturing all rely heavily on data. Operations must work to ensure their data is accurate or they will falter in their steps to modernize.

When discussing the importance of data timing and accuracy, one of the first considerations that comes to mind is the supervisory control and data acquisition (SCADA) system. Timing delays and inaccuracies in signal measurement can be disastrous if not corrected.

An operator controlling a process reacts based on the process information they see. If this information is inaccurate or delayed, the operator could potentially make incorrect process decisions. The more common issue with signal drift will likely be process inefficiencies that drag down the bottom line of the business.

With inaccurate data, there are also problems in finding root cause. When an engineer is troubleshooting an equipment or process problem, the data is the key “witness” they will go to. If the data is not accurate, troubleshooting becomes more difficult – or even impossible.

Maintenance Logs

Naturally, a key component of asset management is the maintenance of that asset. Thus, it follows that the information about past maintenance work is critical to the operation. For long-term understanding and performance of an asset, its maintenance history and future plans should be well understood. This can all be found in a well-kept maintenance log.

The benefits of an accurate maintenance log are rather straightforward. With the history of the asset understood, maintenance technicians can make better decisions on how to go about their work. After seeing repeat issues, the maintenance strategy on an asset can be adjusted to better meet the needs of the operation. Preventive maintenance tasks can be added and worked into the production schedule.

Historically, this was accomplished via paper logs and filing cabinets. A more modern method is via the computerized maintenance management system (CMMS). This digital method is naturally much more robust for many reasons. The data can be accessed from anywhere if needed. There is less paperwork and no chance of losing the paper. Part of the process can even be automated to add work order number, date and person entering information.

Of course, your maintenance logs are not useful if the log data is not accurate. With accurate maintenance history, an operation can get the following benefits:

• the preventive maintenance schedule can be optimized;
• repairs can be diagnosed more quickly;
• spare parts inventory can be reduced; and
• overall maintenance cost decreases. 

However, relying on the diligence of your maintenance technicians can only get you so far. A CMMS can help, and artificial intelligence (AI) can take your recordkeeping to the next level. As AI is the most mentioned disruptive technology today, you should consider its use where appropriate. In this vein, AI has been used to automate maintenance recordkeeping and automatically help to sort and categorize maintenance logs.

Equipment Health

The next level of maintenance is predictive maintenance, where predictive algorithms fed with machine information can give you an idea of when an asset will break down. The algorithms are fed with physical data about the asset, such as vibration, temperature, pressure and so on. Using this information, the operation can plan accordingly and repair assets before they come to an unplanned breakdown. Data timing and accuracy are very important for a predictive maintenance program. The models are fed with real-time data about the equipment, which needs to be accurate for the models to work.

For instance, a predictive algorithm might calculate based on the data from the sensor that a certain part will break down over the next four days. This information isn’t helpful if you need one week to acquire the replacement part. Timing is really important even if you have real-time data.

Smart Manufacturing

All of the general “mega trends” in manufacturing deal with flow of information. Thus, they rely heavily on data – and with “bad” data, they can be rendered useless. Take AI for example. AI has shown to be the most significant predictor to financial gains of any IoT technology today. Therefore, modern manufacturing environments should be planning how they can incorporate AI into their processes.

It is a vital task that the data you analyze for developing any AI is relatively “clean.” Otherwise your AI could guide the process into a non-optimal state. One way to clean up your data is to take it out of a human’s hands. Automating the flow of process information from a machine will make it more robust and more useful. Look into automating production reporting, inventory movement, time tracking and so on.

Another trend in today’s manufacturing is the development of advanced business intelligence (BI) charts and dashboards. These dashboards can offer great insight to the operation – especially down on the shop floor where sometimes the forest gets lost for the trees. Again, if inaccurate data is feeding these BI dashboards, the business could be making the wrong decisions, losing efficiency and profit.

As modern manufacturing moves further into the future with automation and digitization, data is becoming the new “currency.” It is vital that the data driving the digitization of the factory is accurate. Without this important foundation, you are building on quicksand. Specialized software and automation will not always correct the issue. Work toward producing clean, accurate data, and you will be laying a solid groundwork for future endeavors.