Base Maintenance On Data

Integrating predictive analytics into a preventive-maintenance program will reduce downtime by 53% and equipment defects by 79%.

Use these four steps to implement predictive analytics in your preventive-maintenance program.

By John Gaddum, Bosch Rexroth

Industry 4.0 is in full swing and the manufacturing leaders in this era will be companies that are leveraging predictive analytics to create a comprehensive, digital preventive-maintenance program to ensure operations are delivering peak performance. Unplanned downtime, caused by machine failure, has been, and continues to be, a major issue for manufacturers. Unscheduled stoppages often result in lost sales, increased expenses to repair or replace machines, and frustrated customers who may not receive shipments in a timely manner.

Predictive-maintenance programs mitigate those risks by helping plant managers know the condition of their machines so repairs and maintenance can be scheduled in advance to avoid unexpected downtime. The data backs it up. According to a
National Institute of Standards and Technology study (doi.org/10.36001/ijphm.2021.v12i1.2883), companies that integrate predictive analytics in their preventive-maintenance programs reported 53% less unplanned downtime and 79% fewer equipment defects.

The rapidly growing sophistication of automation technology—artificial intelligence, machine learning, and robotic process automation—is contributing to the increasing adoption of predictive-maintenance solutions. So are the increasing sophistication and abilities to monitor equipment wear and alert plant managers to schedule repairs or maintenance before a malfunction affects productivity.

In addition to minimizing unplanned downtime, digital predictive-maintenance programs benefit businesses in several other ways:

Extended machine/equipment lives: Through historical practices such as preventive maintenance, properly maintained machines last longer and perform more efficiently. Plant operators who consistently  monitor asset maintenance and machine usage data are better informed about when to make timely repairs and perform upkeep to extend the lifespan of their machines. Overall equipment effectiveness (OEE) is the common practice for monitoring and measuring the effectiveness of these maintenance programs.

Lower repair and maintenance costs: By proactively identifying and addressing potential production problems before equipment fails, plant managers can reduce the cost of repairs and maintenance. A study by IBM found that the average downtime cost businesses $1.5 million, but that unplanned stoppages were 35% more expensive (ibm.com/downloads/cas/L57KW7ND).

More-efficient service scheduling: Managing and scheduling plant maintenance is a tricky undertaking. Plant managers must coordinate repair/maintenance team schedules and make sure parts and equipment are on hand to complete assignments. Unplanned downtime disrupts that scheduling, which could cause delays in routine maintenance for a business’s other machines. Through real-time condition monitoring, needed work can be identified and carried out during planned production scheduling. Digital repairs and maintenance technology also allow team members to access equipment remotely, potentially cutting service time to minutes or hours instead of days or weeks, particularly if a specialist is needed. The technology has another scheduling benefit: A company’s maintenance staff may spend less time repairing assets, which could help alleviate the shortage of skilled workers.

Improved manufacturing efficiency and product quality: Assets that are operating at peak efficiency are more profitable and generally produce higher-quality products. A comprehensive maintenance program using predictive analytics and cutting-edge equipment sensors can alert managers when machines are not operating at optimal levels.

Simplified monitoring through secure cloud solutions: When machines are connected to a cloud-based network, plant managers can quickly view performance and assess machine condition. The technology can send alerts to managers, indicating when repairs or maintenance should be performed, before a problem occurs. If asset measurements deviate from historical data, self-learning algorithms are effective at spotting those anomalies and generating predictive-maintenance recommendations.

Creating or upgrading a digital preventive-maintenance program that leverages the power of predictive analytics can be challenging. Here are four tips to help minimize the time to launch the initiative and maximize the return on investment:

Develop a comprehensive data-collection strategy. Plant managers often rush to connect all machines and industrial sensors, and then collect lakes of data. That data, however, is of little value if it cannot be organized, analyzed, and presented to managers in a way that allows them to make optimal decisions. A clearly defined data-collection strategy that includes identifying the KPIs that are most relevant to the company’s manufacturing processes, converts raw data into actionable information. It provides managers with valuable insight into equipment health and better maintenance strategies.

Establish a baseline for performance measurement. KPIs are only useful if decision makers have a reference point to compare data. The historical data collected will help plant managers identify equipment use patterns, trends, and anomalies to establish baseline performance levels for their machines. The data can also be analyzed to help determine factors that may have contributed to non-planned downtime due to equipment failure. With a clear understanding of historical machine performance, managers will be in a better position to forecast when machines may need repairs and maintenance. The data will also help them experiment with ways to improve the efficiency of their machines and related equipment.

Prioritize maintenance goals. When creating or enhancing a predictive-maintenance program, it’s easy to fall into the trap of taking on too many projects. Do not try to do everything at once. Start small and move forward. Prioritize your projects and the accompanying goals. Plant managers should also ensure that maintenance goals align with company manufacturing objectives. By looking at the production schedule, downtime impacts, and available resources, decision makers can create an integrated program that better maintains and helps improve the efficiency of mission-critical assets.

Ensure secure connectivity and data protection. Make sure you implement the latest security standards when you connect assets to your IT network. Collaborate with your company’s chief technology officer or cybersecurity partner to identify and address potential vulnerabilities and ensure a seamless flow of information. Protecting your assets with current cybersecurity technology is essential to maintain a secure digital ecosystem for your company. A security breach likely will cause material, financial, and reputational damage.

The factories of the future will be increasingly connected and powered by continued innovation in hardware and software. With a modern, digital, predictive-maintenance program, plant managers can mitigate the risk of unplanned downtime, so the business’s machines are operating at peak, profit-generating efficiency. EP

John Gaddum is Factory Automation Service Manager, North America, for Bosch Rexroth, Hoffman Estates, IL (boschrexroth.com).