How to Develop and Implement Better Maintenance Processes

As the senior maintenance manager at EPCOR’s Gold Bar wastewater treatment plant in Edmonton, Alberta, I feel that two of the most important duties I have are to grow and develop my people and implement and sustain processes. My career spans 29 years in a variety of industries, including mining, forest products, oil and gas, electric power generation and utilities.

Over this time, I have worked as a frontline millwright, planner/scheduler, supervisor, project manager, maintenance manager and now a senior manager. Having worked in so many roles and industries in combination with professional-development training, I have acquired a lot of knowledge when it comes to developing and implementing lean and reliable maintenance programs.

It is not often that the opportunity to truly put your thumbprint on an organization comes along. During the interview stage, it was made very clear to me by the director of Gold Bar that they wanted someone who knew processes. The first day on the job, I was basically told, “You weren’t hired based on your looks, so let’s see what you have in your bag of tricks.” Obviously, I am only joking, but I was made very aware of the great opportunity before me.

On Dec. 18, 2017, I stepped into my current role as senior manager. This article will detail our journey over the past 19 months and how we are moving toward achieving maintenance excellence.

First, I’d like to share the improvement we’ve seen in three key performance indictors (KPIs) from December 2017 to November 2018.

I also want to be clear that I take very little credit for these numbers. As we all know, the most important assets to any company are its people. Without everyone at Gold Bar pulling on the rope in the same direction, this would not have been possible.

Besides people, other key factors include strategy, work management, materials management, processes and support systems. Following is a brief synopsis of what I found in each of these areas when I arrived as well as how we improved things.

People

Steven R. Covey once said, “You can buy a person’s hand, but you can’t buy his heart. His heart is where his enthusiasm, his loyalty is. You can buy his back, but you can’t buy his brain. That’s where his creativity is, his ingenuity, his resourcefulness.”

I was very fortunate that the gentleman I replaced had resolved 95 percent of the human resources issues. I walked into a group of engaged people eager for processes and to be better at what they do.

As I continued to peel layers from the onion, I soon realized there was cross-functional collaboration occurring at certain levels and that there wasn’t a lot of silo work happening.

When it comes to people, the most important thing you can do as a leader is to build trust with your team. Empower your employees to make decisions and accept the fact that they will make the wrong decision from time to time. Pick them up, dust them off and get them back on the road heading in the right direction. Be sure to have a conversation with them so they fully understand that they need to learn from mistakes when they happen.

Henry Ford said, “Even a mistake may turn out to be the one thing necessary to a worthwhile achievement.”

The most enjoyment I get with leading an engaged team is telling them my vision and then getting out of their way. Remove roadblocks and let them soar.

Strategy

When I started at Gold Bar, the strategy was to keep the plant running, which in itself is not a bad strategy to have, but at what cost? The plant is 50 years old, and the culture has been here for some time. I wasn’t going to change it overnight, nor did I know what needed to be changed. I spent the first two months looking at current processes, talking to anyone I could about maintenance, including members of the maintenance team, operations folks, engineering, etc.

I performed a gap analysis and looked at everything from our people and work management to materials management and processes. I was pleasantly surprised that Gold Bar had a very basic form of all the key maintenance processes. They had a skeleton, a framework, but we needed to put some meat on the bones.

To get a better gauge of where we were at the time, I looked at the life cycle of a work order, from the time it was entered into our computerized maintenance management system (CMMS) as a work request until the work was executed and the closeout process completed. In other words, what did our work management look like? Once again, I was pleasantly surprised that there was a work management standard, although it was not very robust.

At least they had given it some thought, and the concept wasn’t foreign to them. The whole idea of work management wasn’t just “get ’er done!” There was a single-line flow chart that had some decision boxes around emergent work, planning and execution, so we had a starting point.

While the work management documents lacked any real substance, maintenance was still being performed effectively. There was great communication between operations and maintenance, and things were fairly smooth.

That being said, there was not much importance placed on protecting the maintenance schedule, and a great deal of negotiating took place on the front line between the operations and maintenance foremen. I had to be careful how this was handled, as the last thing I wanted was to taint that relationship.

Materials Management

As I began to dive into our materials management, I quickly realized there was considerable opportunity for improvement. We had a large vendor-managed inventory of gloves, safety glasses, batteries, tape, etc. that was a free-for-all and costing us lots of money. We had no way of tracking refurbishable spares, so there were squirrel piles everywhere.

Because reliability-centered maintenance (RCM) and failure mode and effects analysis (FMEA) were never performed, we didn’t know if we had the right critical spares in inventory. The culture was to keep inventory low. The planners would buy parts on credit cards rather than creating stock numbers and utilizing supply chain management (SCM). Space was limited in the warehouse, and we were not making effective use of it. We also had no kitting or staging.

Processes

While some processes were lacking in maintenance, safety was well-managed. Change management, process safety management, hazard registry and safe work procedures were top notch. Again, I was pleased that we had an engaged group of employees who took safety seriously. It showed in their statistics, with only one recordable injury in 2017 and one in 2018. Considering the level of engagement, I knew it would not be difficult to develop the processes.

Support Systems

Our CMMS program has a lot of horsepower, but we were only utilizing a small portion of its functionality. We also had some very convoluted ways of getting information into the system after the work had been executed. The tradesmen would handwrite notes onto the paper copy of the work order.

The foreman would close the work order. The paper copy would be sent to the maintenance admin, who in turn would reopen the work order and type the tradesmen’s remarks into the CMMS.

We were also doing some condition-based monitoring. Oil analysis was being performed offsite. Vibration readings were collected by the millwrights and analyzed by engineering, but no reports were created or data correlated. There was no effort to identify bad actors or develop data such as mean time between failures (MTBF).

Below are some of the findings from the gap analysis:

• Great workforce – kept the place running

• Old-school mentality with respect to following work management processes

• Weekly maintenance schedule was created the Thursday (three days) before it was to be executed

• No formal look-ahead for maintenance

• No cleanup of maintenance backlog

• Break-in work wasn’t based on equipment criticality

• No schedule protection

• Poor data input into the CMMS

• No formal strategy for maintenance

• Too many parts purchased on credit cards rather than through SCM

• Poor inventory management, too many squirrel piles

• No kitting or staging

• Warehouse was open and basically an honor system

As mentioned previously, it took 50 years to reach this point, so we were not going to turn the ship around overnight. We needed to set small, measurable goals and to walk before we ran. Any time you implement something new, it is vital to get in front of the change. Communication is of the utmost importance.

You must communicate the change to your team as early as possible, prior to implementation. Let it sink in, make yourself visible and available to answer questions, and dispel any fear. Make sure your leaders are on the same page, because if it is not important to the ones who lead, it won’t be important to anyone else.

Everyone reacts to change in a different way. One of the most important things about change is ensuring you have a built-in sustainment piece. Follow the W. Edwards Deming model of plan-do-check-act. Monitor the change closely and adjust accordingly. Without the sustainment piece, many change initiatives fizzle out and become a flavor of the week. 

Picking Low-Hanging Fruit

The cross-functional collaboration and respectful, constructive debate I encounter on a daily basis are not found in too many places. These people have a passion to do the right thing. It all begins at the daily core team meeting at 8:30 a.m. This is a cross-functional meeting with representatives from operations, plant engineering, plant controls, automation, project engineering and maintenance.

All work requests that have been entered into the CMMS for the past 24 hours are reviewed and prioritized based on the priority designation entered in the work request. The date that operations would like to have the equipment back in service is entered on the request. The scheduled date for execution will be reviewed once planning is completed and the job is kitted and staged.

The challenge was getting people to focus on the equipment’s criticality. If the asset is not critical to the plant’s operation, it should not break into the schedule. I had to reinforce with the maintenance foremen that they needed to protect the schedule. Immediately, we started to see changes to the break-in work and schedule compliance KPIs.

Planning and scheduling were next. How confident can you be that the schedule will be completed when you develop it on a Thursday the week before work is to be executed? Without ensuring all jobs were kitted and staged, I had little confidence. However, they still had 70 percent schedule compliance, which I thought was reasonable.

I held a meeting with my leadership team, and we implemented a five-week rolling schedule. We now lock the weekly maintenance schedule 10 days prior to execution. While the schedule is locked 10 days in advance, it is reviewed at the weekly scheduling meeting to ensure plant priorities haven’t changed and that all jobs have been kitted and staged. If all the parts required are not kitted for the job, it is removed from the schedule.

We also instituted parameters for how we manage emergent or break-in work. This was one of the key changes to increase our preventive maintenance (PM) performance. In the core meeting, if there was emergent work, the first thing to be pushed from the schedule was low-priority corrective maintenance, then PMs and finally work that required more extensive isolation.

Along with the professional-development training I have completed over my career, I’ve also done a lot of work with lean systems. So, whenever I am looking to make changes, I do so from a lean mindset. I’m always trying to drive waste out of the process and am a firm believer in only doing value-added work.

The next area we dove into was preventive maintenance. There was very little maintenance performed based on condition. The majority of PMs were time-based. I explained the P-F curve to our guys and said that performing PMs is vital in moving us from firefighting to fire prevention.

Frequency is also very important. If we are doing weekly PMs but not seeing any changes, we may need to free up some man hours by moving the frequency out to two weeks and so on. Over the last half of 2018, we modified more than half of our 1,400 PMs. We saw a dramatic increase in PM completion and the amount of work requests that reflected “corrective work from PM.”

Over the course of the last nine months of 2018, we focused on developing our work management process. It is now quite detailed. Our motto for 2018 was “doing the right maintenance at the right time.” With that mindset, we have been very successful. In retrospect, what did we really do? We looked at work requests from a different lens based on the equipment’s criticality to the operation.

We prioritized the way work was executed so there was an emphasis on completing PMs. We made changes to the weekly scheduling meeting to look ahead five weeks and locked the weekly maintenance schedule 10 days prior to execution. We implemented kitting and staging. Everything that was on the upcoming maintenance schedule had to be kitted and staged.

We now have bi-weekly schedule-compliance review meetings with the maintenance foremen. Once we started these meetings, we saw quite an increase in schedule compliance. The maintenance foremen knew then that they were accountable for the schedule, and they owned it.

We also created a sub-inventory class in our CMMS so refurbishable spares could be entered into stock with an associated value that would be charged to an asset when it was used.

In addition, people were held accountable. I am a firm believer that everyone has a bucket of work to do, from the person on the end of a hose to the CEO, and you should be working in your own bucket.

Where to Go From Here?

Now that we have the basics down, we will slowly add new things to continually improve. We provided training to all maintenance employees so they can enter all their findings directly into the CMMS and complete their tasks. We are starting to implement failure codes in the CMMS so we can track why equipment is failing and trend bad actors.

We have chosen four maintenance employees to be our condition-based monitoring group. This year we will train them in vibration analysis, ultrasonic testing, thermography and tribology. The goal is to have a competent and self-sufficient condition-based maintenance team in four years.

In December, we conducted a three-day RCM training course that 17 Gold Bar employees attended. This allowed us to review all our systems using the RCM model and ensure that we have the right strategies for our assets, that they are assigned the correct criticality and that we have the critical spare components in inventory.

Our strategy has changed. By the end of 2021, Gold Bar will transition from a reactive and preventive maintenance organization to an on-condition and predictive maintenance department. Sharing the new strategy and vision with the team and all the members of the Gold Bar family helps to keep everyone focused. If they know the goal, they will help us build the plan.

This article was previously published in the Reliable Plant 2019 Conference Proceedings.