Downtime, and especially unplanned downtime, is expensive with some cost estimates exceeding $1 million an hour. When a line is down, overhead expenses continue but no product is being produced and no profit is being generated. If the downtime delays deliveries, customer satisfaction plummets.
Common causes of downtime include general wear and tear, operator error, product changeover, unexpected component failure, maintenance schedule mismanagement, machine programming issues and machine consumables problems, according to Packaging and Predictive Maintenance, a recent report published by PMMI Business Intelligence, a division of PMMI, The Association for Packaging and Processing Technologies.
To maximize uptime, machines and systems must run reliably, consistently and at required speeds. Proper maintenance is crucial. Preventive maintenance forestalls breakdowns through actions such as lubrication, tightening components and replacing worn parts. Such tasks are performed according to a regular schedule in an effort to minimize unexpected failures and avoid unplanned downtime.
Predictive maintenance takes this concept a step further by using digital tools to monitor and analyze asset behavior, predict asset behavior and perform repairs proactively to prevent unplanned downtime. This requires equipment that can monitor its status by collecting data about its performance such as run time, voltage (especially for motors), speed, pressure, temperature and vibration and provide a warning if attention is needed.
Predictive maintenance is one of four key priorities identified in another recent PMMI report, Challenges and Opportunities for Packaging and Processing Operations. Avoiding downtime and preventing product loss are the major drivers pushing process manufacturers toward implementing predictive maintenance. Other benefits include longer machine life and a reduction in parts requirements because replacements are installed as needed rather than on an arbitrary schedule.
As a result, there is a huge demand for technology that increases machine reliability, and interest in predictive maintenance continues to grow. According to the report, 43% of consumer packaged goods (CPG) companies currently use predictive maintenance and 45% are planning to implement it in the next three years. Of those using the technology, many firms are still working on installing it throughout their operations.
CPG firms using predictive maintenance rely primarily on three tools: thermography, full equipment monitoring and CMMS (computerized maintenance management system) software. Other areas of interest include vibration analysis, machine health monitoring sensors, parts room setup and organization, oil monitoring analysis, risk and reliability software, machine-level fault codes, tracking hours of use until downtime, outage planning and scheduled TPM (total productive maintenance), risk minimization, best practice and mean time between failures data.
The most successful applications of predictive maintenance depend on receptive management and carry the most impact when deployed in association with critical assets and expensive components with long lead times. Predictive maintenance practices are particularly useful on 24/7 lines and help eliminate bottlenecks. Key drivers include delivery of actionable information, ease of application to new and legacy equipment, good ROI, scalability and the ability to collect the right data.
However, collecting the right data often depends on remote access to the machine. Historically, this has been a tough sell to CPG companies due to cybersecurity concerns. However, the COVID-19 pandemic, which limited, or even eliminated, in-person service calls, forced many CPG companies to reconsider remote access and shifted attitudes. “Remote access is more acceptable now,” Adrian Lloyd, CEO & research director of Interact Analysis, told the audience during The Future of Automation, a webinar, sponsored by PMMI in December 2022. “CPG companies have realized remote access is doable, and there are benefits associated with it.”
Implementing predictive maintenance depends on two categories of technology: hardware, typically smart sensors, and software/analytics (often hosted in the cloud). PMMI’s Packaging and Predictive Maintenance report notes new business models are needed to ensure that predictive maintenance delivers on its promise of optimizing equipment performance for CPG companies while ensuring original equipment manufacturers (OEMs) generate the revenue they need to stay in business. One promising business model, Machines as a Service, or MaaS, replaces the outright sale of a machine with payments based on output. With maintenance as an integral facet of this business model, there is potential to minimize downtime and maximize machine lifetime and deliver advantages to both end user and OEMs.
Another tool to help processors establish a maintenance program is the Asset Reliability Roadmap for the Consumer Products Industry. Developed by PMMI’s OpX Leadership Network, ithelps CPG companies and OEMs understand common definitions; outlines key performance indicators (KPIs) related to people, operations and finances; provides useful calculations; and delivers the leadership guidance needed to develop an asset reliability initiative. Both CPGs and OEMs gain an understanding of the necessity of providing a solid business case for calculating maintenance program costs. This cooperative effort can yield significant improvements in overall equipment effectiveness by focusing on opportunities to reduce the impact of planned and unplanned downtime.
“When developing this tool, our team of CPG and OEM maintenance experts shared their individual and varied experiences, insights and leadership guidance to identify and define a number of KPIs that evaluate asset reliability,” said Bryan Griffen, director, industry services, PMMI. “As a result of this collaboration, the OpX Leadership Network is excited to provide the industry with a unique, effective tool that will help businesses quantify the total value of asset reliability to the organization.”
What comes after predictive maintenance? Prescriptive maintenance. This concept, which relies on machine learning, is in its infancy but represents the next level in the maintenance hierarchy. Instead of monitoring machine status and recommending when to perform maintenance, prescriptive maintenance uses machine learning to not only determine when an asset will fail but also how to fix it. After the work has been completed, monitoring continues to ensure the action taken eliminated the problem, and the tool works out possibilities for the next improvement.
Opx Leadership Network tools, including the Asset Reliability Roadmap, are available for free download at www.opxleadershipnetwork.org. Another resource, PACK EXPO Las Vegas (Sept. 11-13, 2023, Las Vegas Convention Center), will showcase many machines equipped with downtime-mitigating technologies. For more information, visit packexpolasvegas.com.
Jorge Izquierdo is vice president of market development for PMMI, The Association for Packaging and Processing Technologies. He oversees PMMI’s market development plans, research and programs for strengthening the competitiveness of North American suppliers of packaging and processing equipment.