Taking the initiative: achieving greater uptime and productivity with predictive maintenance 

With the huge annual cost of unplanned downtime to industrial manufacturers, it is vital that businesses ensure maximum asset availability and achieve the highest possible production efficiency. Predictive maintenance, powered by data analytics and IoT, holds the answer. Kevin Bull, product strategy director at Columbus UK, explains

From continued efforts on mega-trends such as sustainability to rising ingredients such as CBD-infusions, 2019 is likely to be no less a year of change than those preceding it. Callum Tyndall examines what the year ahead looks like for the food industry.

Image Courtesy of Michelle Lee Photography /

The food manufacturing sector often faces significant challenges, including stringent government regulations, a harsh production environment and the need for maintenance of complex equipment. Any equipment downtime threatens production losses, a negative impact on food quality in the value chain, and damage to the overall brand image. Although a set amount of scheduled downtime is a necessity for operations such as changeovers, cleaning and tool changes, unplanned downtime can have a significant impact on business-wide operations and incur significant cost. In order to reduce this downtime frequency, a manufacturer needs to be able to ensure outstanding performance of equipment, now made possible through technology-enabled close monitoring and a shift from reactive to proactive maintenance. 

The perils of poor equipment maintenance

Without a proactive approach to equipment maintenance, unplanned downtime is inevitable. Recent research estimates that unplanned downtime currently costs industrial manufacturers up to $50bn annually. Disconnects in the supply chain lead to lower production, missed deadlines and risk incurring penalties. On top of the impact to production levels, for the food industry there is also a threat of knock-on effects to other areas of business.

When food manufacturing equipment is outdated or on the verge of a breakdown, it is prone to product contamination. Unreliable equipment can also produce lower quality products that attract penalties and expensive recalls. If a food manufacturer begins to miss deadlines and produce unsafe or poor quality products, it could seriously tarnish the overall brand image.

It is easy to see why equipment maintenance has now become the most critical compliance factor for food manufacturers to adhere to. In the US for example, the FDA’s mandate includes equipment maintenance as one of the risk-based preventive measures in the Current Good Manufacturing Practices (cGMP). The United States Department of Agriculture also considers equipment maintenance a key function of cGMP for any food safety management system.

Predictive maintenance: the gamechanger

Predictive maintenance (PdM) harnesses the power of advanced technologies such as IoT in real-time, to track, monitor and gain deeper insights into food manufacturing operations. This includes the performance, availability and quality of equipment mandated by overall equipment effectiveness. It also keeps a check on the risks of unplanned downtime and minimises planned downtime. PdM collates data from critical equipment sensors, ERP systems, computerised maintenance management systems and production data to define and deliver prediction models in an analytical fashion. This data aids in predicting equipment failures and assists with taking a proactive approach to addressing these.

Traditional preventive maintenance systems are often time-consuming, and it is quite difficult to get a clear picture of how, when and where an error can occur. PdM however, powered by data analytics, is instant. It races ahead of traditional systems, pulling data on asset utility value, availability and reliability. Using machine learning, equipment maintenance managers can detect deviations and equipment failure. This helps them proactively take better decisions that further translate into increased productivity.

On top of this, when you deploy IoT-connected sensors, you can gather data from materials and equipment in real-time – no matter how long or how short processes are. This means you can keep close tabs on temperatures, sugar content, alcohol levels, carbon dioxide levels, and other aspects of in-progress production. When these data findings are analysed, you can forecast the completion of these processes. You now have accurate and real-time data to help optimise production planning and efficient use of production capacity, enabling customers to get their desired products on time.

Share this article