You run a manufacturing business. You are the owner, CEO, COO, manufacturing director, divisional managing director, plant manager, or another equally responsible role. Every day, whether you commute to the factory daily, work from home and monitor the operation remotely, or wait for the phone to ring with unwelcome news, you have the same worries: will we make all our commitments today? If you have implemented a PdM system you would have peace-of-mind and not have to worry.This is because we know that there are so many things that can go wrong on the manufacturing floor that any day without a major crisis is a good day. And because you know that, on average, downtime can cost manufacturers up to $260,000 per hour, according to analyst firm Aberdeen Research.

Good News, Bad News

The good news is that most things that can go wrong in a manufacturing process are under your control. You can achieve your desired outcome by identifying, measuring, and improving all your unique processes.

The bad news is that one area not yet under your control is a machine failing and stopping all or some of your processes. Furthermore, you know that Murphy’s Law is working on every production floor and that a machine will fail when it creates the most chaos, confusion, stress, and problems for you and your team.

Fortunately, new and legacy production machines can be outfitted with effective dMP systems!

Based on data presented in the Packaging and Predictive Maintenance report from the Association for Packaging and Processing Technologies, Inc., this chart shows where their CPG machinery OEMs stand on implementing PdM. Approximately 75% of the OEMs are in some stage of implementation!

 

Medium_Screen Shot 2022-01-14 at 1.25.45 PM.jpg - a few seconds ago

 

What Is Predictive Maintenance (PdM)?

PdM is a technology-intensive system that continuously or periodically measures physical parameters on a machine, transmits the measurements to a central computer, and analyzes the data with artificial intelligence machine learning software. When the software determines that a hardware failure will likely occur shortly, it warns the equipment operators and maintenance people so they can schedule a repair when it will have the least impact on production outputs. While this sounds complex, the use of PdM is becoming routine. The vendors are taking care of selecting and installing all the technology, and all the machine owner has to do is provide some money and people who will learn how to identify and prioritize equipment problems.

Instead of working with an equipment vendor, companies using rotating equipment can buy a hardware kit from Amazon, connect a data stream to AWS (Amazon Web Services), and use their PdM software (Amazon Monitron Can Empower You to Easily Implement Predictive Maintenance). However, they still have to provide the people who will repair the equipment.

Data and People Aren’t Enough to Turn PdM into Uptime

Turning PdM data into equipment uptime requires a strong capability in four unique areas. Those areas are:

  1. Predictive maintenance data
  2. Spare parts
  3. Qualified people
  4. Knowledge to help qualified people prevent a problem
Medium_Screen Shot 2022-01-14 at 1.24.07 PM.jpg - 2 minutes ago

In 1982, Eli Goldratt and Jeff Cox published The Goal. This novel takes place in a manufacturing organization and explains how the plant manager works his way out from under a delivery crisis. The ideas in the book eventually became the Theory of Constraints (TOC), which is a management paradigm that sees any manageable system as limited in successfully reaching more of its anticipated goals because of a small number of limitations. Through a focusing process, TOC identifies the presence of at least one constraint in every scenario and rearranges the rest of the organization around that element. Related to the well-known phrase “a chain is no stronger than its weakest link,” TOC sees businesses as vulnerable because the weakest component of a system can always break, negatively affecting the rest of the process and the resulting outcome.

What this all means is that a constraint on manufacturing outcome can result from a lack of information about an impending failure, a missing spare part, an unavailable mechanic or repair technician, or have all three but no idea about how to prevent this specific problem. As soon as the problem is cleared, a new constraint will emerge.

How to Cover All the Bases

There are a few ways to ensure that you will have all the required elements when a problem occurs or will likely occur soon.

PdM Warning

Develop or purchase an IT architecture so that any PdM system can collect data and integrate it into a plant-wide view of all the equipment you monitor. When buying the equipment, include PdM sensors and interfaces to integrate into your plant-wide system. For critical legacy equipment, buy PdM upgrades from the OEM or other independent system providers. Make sure you can integrate the information into the plant-wide system.

Spare Parts

Work with each manufacturer of your equipment, whether PdM-monitored or not, to provide a complete set of spare parts and repair or replace defective parts. Turnaround availability and replenishment time should be specified and monitored, which may involve an annual fee or a predetermined replacement cost.

Qualified Technician

On-site response time is critical, so the first line of defense is frequently a qualified plant maintenance technician. To minimize downtime, you will need coverage whenever the equipment is running. An alternative or backup is a local technician working for the OEM or working for you on a temporary basis. The least desirable alternative is an OEM technician who must travel a significant distance to get to your facility.

Once, I was the production manager for an electronics company. Our unique machine produced a unique assembly used in over 75% of our products. Our machine owner was trained at the OEM’s factory and had a backup who was trained in-house. An additional backup was three manually operated stations that, when combined, could produce about 50% of the automated machine’s output. We had a WIP stock of five days of usage.

One day, the automated machine failed, and we spent the next three days troubleshooting it with the OEM’s telephone assistance. I finally bit the bullet and asked the OEM to send one of their technicians to work on our machine. All was good… except I was located 30 miles west of Boston, and the OEM’s tech was located at the factory in Grand Rapids, Michigan. It took him longer to get to our facility than it took him to repair the machine—ugh!

Available Knowledge

There are two kinds of knowledge: explicit (institutional) and implicit (tribal). Here is how Dozuki defines each type of knowledge:

“Explicit or tangible knowledge is the concrete information that comprises essential information and data. Things like standardized procedures and safety protocols are documented by necessity, and are crucial to operations. This knowledge is easy to store and pass between people because it exists in the form of documents, records, or reports.

Implicit or intangible knowledge includes personal stories, skills, and intuition-based learnings that are accrued through experience, in-person training, or mentorship. This type of knowledge is more difficult to communicate and often remains siloed or lost. A bulk of tribal knowledge falls in this group.”

The knowledge must be available and appropriate for the user. This means that while a work instruction (explicit knowledge) can be stored on a file server, it should exist in enough versions with different details so the experienced technician can get a high-level view of what has to be done, and once reminded, she will know exactly how to do it. On the other hand, an inexperienced technician needs a detailed step-by-step set of instructions with enough pictures to ensure she can confidently fix the problem. Also, ideally, each version should be available in enough languages so anyone can use them.

All knowledge should be current. If a design changes, the instructions should be amended to include the new version without causing confusion.

Looking Ahead at PdM

Implementing a successful PdM system requires a lot of work, but it is worth the effort, because you need to get those four areas right and then get your team to accept this tool.

In today’s manufacturing environment, it is necessary to have an operational PdM system installed on your critical equipment. However, that is not sufficient to ensure minimum critical equipment downtime. You must also have access to well-trained service technicians, a complete supply of spare parts, and specialized information to help your team solve all problems.

Here is a related article about PdM.

About Middlesex Consulting

Middlesex Consulting is an experienced team of professionals with the primary goal of helping capital equipment companies create more value for their clients and stakeholders. By focusing on our services, customer experience, manufacturing, and engineering strengths, Middlesex Consulting continues to provide superior solutions to meet its clients’ needs. If you want to learn more about how they can help your organization improve its revenue creation processes, please contact them directly via email or check out some of our free articles and white papers here

Image by Gerd Altmann from Pixabay