Counting the Cost of Asset Failure
SummaryJustin Eames, Senior Manager, APM Solution Consulting, Aspen Technology Inc, looks at asset failure and explains how predictive analytics could help manufacturers avoid expensive over-maintaining
- Author Company: PharmiWeb.com
- Author Name: PW Editor
- Author Email: email@example.com
- Author Telephone: +441344851506
Counting the Cost of Asset Failure
Justin Eames, Senior Manager, APM Solution Consulting, Aspen Technology Inc, looks at asset failure and explains how predictive analytics could help manufacturers avoid expensive over-maintaining
Asset failure is simply not an option for pharmaceutical companies today. Interrupting the production and supply of drugs is likely to mean they lose out to the competition and will negatively impact on customer loyalty. To prevent any possibility of this happening, many manufacturers across the sector indulge in expensive over-maintenance.
This process can involve taking machines offline for lengthy, and costly, periods during which time they are typically cleaned and dismantled, and replacement parts are added. While it might protect the batch, this kind of usage-based maintenance typically equates to lost production, even leading to manufacturers investing in additional assets at great expense or having to incur CAPEX to “buy” more plant volume.
The consequences of asset failure
The situation means companies are faced with having to ensure the continuity of supply while using less equipment, which is driving a greater focus on the equipment’s reliability. In the past, where customers had typically invested in over-maintenance or excess capacity and were looking to avoid wastage, they were only able to gauge whether production was more reliable through their own applied intuition.
Drugs are far from cheap - typically, they can be worth hundreds of thousands or even millions of dollars per batch. An asset failure will usually in some way corrupt or contaminate the entire batch, creating waste, loss of valuable product and supply disruption.
There are common types of asset failure, such as rotating machinery like pumps or spray/freeze dryers, that are used by most pharmaceutical manufacturers. Often assets, such as mixer seals, are replicated across numerous assets and sometimes multiple sites. By finding the solution for predicting failures on that seal, digitalisation can be used to apply it to the rest which has a huge impact in terms of overall reliability. It is very much the ‘build once, apply to many’ principle in action. So why hasn’t there been a bigger shift in manufacturing towards more digital tools, maintenance and analytics?
Industry challenges of moving to digital
By necessity, the pharmaceutical industry already has well-established, often complex validation processes (albeit more manual than digital) in place to be able to manufacture the medication. This means customers tend to be reluctant to take on additional “evaluation pain” in a bid to achieve a more rapid time to value. This thinking may not be ideal, but it is delivering the required validation to produce the drugs.
On the positive side, digital products don’t directly have a physical impact on the drug. Ultimately, they give advice that someone will act upon, under normal procedural controls. As a result, the systems can be easily implemented with relatively little validation demands. This simplicity is an issue that pharmaceutical customers don’t really appreciate. The implementation of these digital systems doesn’t necessitate an overhead in terms of validation; however, they do require a mechanism to provide advice for someone to make a decision on.
Looking ahead, there is now a real recognition that digital technologies will drive value for pharmaceutical manufacturers. The sector is looking at capacity and processes – finding new ways to become leaner and more efficient; predictive analytics is one of the key areas that digitalisation is enabling this.
The power of predictive analytics
Seeing into the future with the level of accuracy that can now be provided by cutting-edge predictive analytics is a potential game changer for manufacturers. Depending on the length of the batch, in terms of maintenance users often see predictions of many days or even weeks ahead. This provides the ability to predict asset degradation and failure in advance of an impending breakdown.
Predictive analytics provides the ability to make decisions that can not only minimise cost and disruption but that can also protect public health by ensuring continuity and resulting quality of drug supply. Its two key benefits for businesses are being able to plan ahead for the remedial actions that they need to do, while avoiding those other activities that the technology reveals are not actually required.
The analytical tools behind this capability can enable manufacturers to get more capacity out of their existing equipment. By looking at prescriptive predictive solutions it’s possible to run for longer with less maintenance. As strange as it may sound, a large proportion of failures are induced by maintenance. It turns out the old saying, “if it ain’t broke, don’t fix it” is key to the avoidance of costly over-maintenance.