Rally to the Customer
Armstrong World Industries makes billions of square feet of ceiling tiles every year, shipping to locations all around the world. Most of these tiles are white. The remainder are off-white or a few standard colors. If a customer wants non-standard colors, until recently, Armstrong used an outside vendor.
In 2017, the vendor was getting behind on orders from Armstrong. I was asked to join a team visiting the vendor to find out what was causing the delays and to identify options. We took a guided walk of the facility. I could see things weren’t well organized and processes were backed up. When the rest of the team attended meetings with the vendor leadership, I decided to stay out on the floor and find out what was really going on. I spoke with a number of operators, who were very open to discussing the current situation. It appeared things were about to change in a big way and the managers weren’t paying much attention to the day to day operations. Every order was behind.
Overlooking Operations
When the other meetings were over, I reviewed what I had observed with the team. They found out the vendor was about to expand capacity and new product offerings. They weren’t spending much time or attention on the day to day operations. Moreover, the factory was to be shut down for a number of weeks during the expansion project. Some of the new equipment used technology the vendor had no experience with. I knew we wouldn’t be able to rely on the vendor in the short term and possibly long term. We needed to come up with another way to provide non-standard color tiles to our loyal customers. If not, they wouldn’t stay loyal to us.
When I returned to Armstrong’s corporate center, I reviewed my findings with my sponsors. We agreed we couldn’t rely on the vendor to meet customer demand in a timely fashion. We had to come up with another solution. I was tasked to find the solution quickly. Working with my counterpart in Marketing, who was responsible for the relationship with the vendor, we agreed we would find other options to produce non-standard color tiles and then reduce the demand to the vendor in steady increments. He took the vendor relationship and product management responsibilities. I took the production, operations, and logistics responsibilities. We took a walk into the Pilot Plant to see what might be possible.
The Pilot Plant was a series of disconnected operations designed for scientists and engineers to experiment, test, and develop new products. It was not a manufacturing facility, and was not designed to be. Upon review of the operating permits, we determined we could produce a limited amount of material on a temporary basis. Now we had to figure out how to produce products safely and effectively with our available resources.
The Customer Comes First
One Monday morning I visited with the technicians who were responsible for operations in the Pilot Plant. I reviewed the situation. If we didn’t figure out a way to produce thirteen non-standard colors of tiles, we would lose valuable business from our loyal customers. Most of the technicians weren’t keen on trying to turn the Pilot Plant into a production facility, but they realized how important it was to keep customers happy. Some of the technicians volunteered to help me determine the best way to connect and run the disconnected processes in the Pilot Plant.
We decided to focus on the edge cutting, edge painting, and tile painting processes in the Pilot Plant. The prior processes, which included forming the board and cutting it to size, could be provided by one of the many manufacturing facilities in the Armstrong network. These facilities were set up to make large production quantities and these early process steps were efficient for them. The edge cutting and painting steps would differentiate the tiles and had to be well controlled, with low amounts of inventory. The large manufacturing facilities were not set up to run small production runs efficiently.
There were many separate areas in the Pilot Plant, but the key areas that we needed were located very close to each other: paint mixing, edge cutting and painting, and overall tile painting and paint drying. These areas were not connected however, and we didn’t want people to have to physically transport the tiles through the line. Our first priority was to build and install temporary conveyers to flow materials through the line from beginning to end. Luckily our technicians were very creative and within days they had cobbled together the conveyers we needed.
The next step was to determine how to process the tiles safely and productively through the entire connected line. We had to determine the proper amount of people, the pacing, the supply of materials to the line, and the handling of finished product.
Timing is Everything
I took a small team of technicians and we decided the best way to design the process was to try it in many controlled experiments. We gathered a bunch of scrap tiles to test and started loading them into the beginning of the process. “Be the tile” was our rallying cry. “Let’s walk the tile through the process and see what happens to it,” I said. We started at the beginning, the edge cutting and painting process. Three technicians were the loading crew. Two technicians positioned themselves at the end of the process as the unloading crew. They started putting one tile on the line at a time and quickly realized they couldn’t keep up with the speed of the line without adding one more resource, who brought full pallets of tiles to the line on demand. Think Lucy and Ethel on the famous Candy Factory episode of I Love Lucy.
Once we added the resource, the technicians were able to comfortably put a tile on the line every four seconds. They were very proud of themselves, but we quickly realized our short test wouldn’t be sustainable over long work shifts, with different people doing the work on the line. After much discussion, we agreed on loading a tile every five seconds. It doesn’t sound like much of a difference, but it could impact production output by 25%. More importantly, it would ensure we were working as safely as possible.
Having confirmed the pacing at the front end of the process, we verified the other parts of the process could be optimized to this new rate. Then, we designed the quality control process and packaging station to meet the pacing requirements safely. We then built a logistics plan and operations schedule. Reaching out to the total Armstrong organization to get volunteers to help staff the line during these critical production runs, we had many volunteers, from administrative staff to vice-presidents to accountants. Everyone wanted to help out and ensure we didn’t disappoint our customers.
The next few months, we produced the non-standard color tiles and met almost every order on time. More importantly, people who hadn’t worked together previously got an opportunity to do something special. We all rallied to protect the interest of our customers and the integrity of the business. It made our bonds stronger as an organization. It wasn’t apparent that we were producing these tiles in a different way and eventually we were able to outsource the orders to an Armstrong manufacturing facility. The demand for these non-standard color tiles and customer loyalty continues to grow. I am convinced it’s because of the special people who put the customer first and their normal duties second.