Valmont Industries

It Just Can't Be This Simple

Valmont Industries produces steel light structures for the utilities and construction markets. The company has two basic products - small pole applications for highway and street lighting and traffic signals, and large pole structures for power line transmission, high mass lighting, sports lighting, and communication towers.

Before 1987, Valmont used an MRP system to assist in scheduling the shops. They launched jobs for firm customer orders and for forecasted requirements. Valmont also used daily summaries that provided total hours scheduled through the work centers to give them the ability to see the hours that are scheduled through the shop. Through control listings, they were able to predetermine and set batch quantities to minimize set-ups and to maximize efficiencies through the shop.

Valmont's BOM structure was fairly simple. The routing files were based on and established by IE calculations. They have approximately 7000 to 8000 active part numbers, 40 call centers, and a computerized inventory tracking system.

Based on MRP, all shipping-level items were scheduled with a finish date of Monday of the shipping week. MRP would then backwards-schedule the shop dates. Due to batches starting by the week, Valmont had a lot of hours starting at the end of the prior week and they generated schedules trying to keep capacities level throughout the entire shop, while maintaining predetermined levels of efficiencies and productivities on all work centers.

Valmont paid very close attention to the fit-up and weld areas of the shop, which were scheduled to total capacity. When scheduling these areas, they were scheduling firm customer orders and forecasted requirements.

Although these systems were in place, Valmont experienced a lot of problems with MRP. They continuously faced an enormous expediting task due to end of the week starts. Priorities were set on orders with the least amount of shortages rather than on orders that were the latest in the shop. Past due orders were growing continuously and rescheduling became the norm rather than the exception. Each week that passed, the same thing happened.

Also, due to the predetermined batch sizes, large quantities of parts were moved from one machine to the other in the parts processing areas, which in turn caused big queues to build in front of the machines. This caused problems getting parts down to the fit-up areas so the welders could have something to work on Monday, Tuesday, and Wednesday of the following week.

In 1987, the staff was told to read The Goal. The consensus was that The Goal must have been written about Valmont.

The first step they faced was deciding if what they read in The Goal could really work in Valmont Industries. Would Valmont have to eliminate existing systems and start all over, or could they somehow interface the two? They were afraid if they made any changes they would shut down the shop - but they also knew something had to happen - some change had to take place because there were too many problems.

While implementing TOC in the shop, Valmont had to develop a new mindset. They came to understand that set-ups, productivities and efficiencies didn't mean anything unless they were tied to the constraint.

All lot sizes and minimums were reduced to one-to-one. Forecasting was totally taken out of the system. After doing this, Valmont's process time was significantly shorter than the customer quoted lead times, which eliminated the need for finished goods inventory.

The next step was to implement Drum-Buffer-Rope. The drum is the finite scheduling of the system's constraint. The buffer is time protection for the constraint against any disruptions. The rope is the releasing of raw material to the shop.

Valmont now launched orders in the constraint based on the start date. They allowed MRP to forward-schedule the finish date, and scheduled the subordinate areas based on the finish date and allowed MRP to backwards-schedule the start date. The biggest hurdle they faced was realizing that it was okay for the non-constraint machines to sit idle.

It became very common to see workers in the subordinate areas painting equipment or doing preventative maintenance, as long as there were no buffer holds.

Valmont continues to measure the constraint based on traditional efficiencies, productivities and performance to the schedule.

Once this was implemented, Valmont realized that if these changes were really going to work, the shop employees had to be educated in TOC. The shop employees were brought in small groups to discuss TOC and run several of the dice game simulations (from The Race). Valmont expected negative feedback, because quite naturally, many people do not adjust well to change. Instead, there was a lot of positive feedback that helped them proceed with implementing Buffer Management in the subordinate areas.

After implementing TOC, Valmont realized significant improvements:

• Immediate reduction in WIP
• Small queues in front of machines, rather than big queues
• Forecasting requirements began to disappear
• Employees were actually going to other work centers to get parts to bring back to their own areas to work on - rather than the parts being pushed through the system the MRP way, parts were being pulled through the system
• End-of-the-month crunch vanished
• Throughput increased significantly
  1986 shipments - 6.5 million
  1996 shipments - 46.5 million

Since Valmont had success with TOC, they had to ask themselves, "What's the next step? What is the process on ongoing improvement?" They found out that if the focus and basis of all business decisions remained on Throughput, Inventory, and Operating Expense (T, I & OE) and if they continuously applied the five focusing steps of TOC, Valmont would continue to be successful.

What Valmont has done the last ten years is before elevating the strategic constraint, they review their CCR Buffer Analysis to determine the top two major Region 1 buffer violators. Once identified, Valmont decides what has to be done to make improvements - maybe buy additional equipment, add manpower, make process improvements, additional facilities - whatever it takes to eliminate the major buffer violators.

They also look behind the constraint to determine which cost center has chronically given them problems after the product leaves the constraint, and then apply the same logic - what can be done to solve this problem?

Valmont has continuously done this - over and over and over again - for more than ten years. Can it be just this simple? Yes!

These results were presented by representative of Valmont Industries at the Jonah^SM Upgrade Workshop in March 1997. The presentation is available on video (#JSA-12).