Begin with the Basics
“Lean” and “Six Sigma” have become popular but while many have heard the titles, some may not know the differences or significance, so it’s best to begin with their basic definitions. Lean Manufacturing (Toyota being the most famous user) focuses on the elimination of waste. Examples of categories of waste are: overproduction, poor quality, and waiting. By eliminating waste, design and production lead times are reduced; fulfillment rates, cash, capacity and customer service / satisfaction increase.

Similarly, the principles of Six Sigma (GE being the most famous user) are to focus on reducing process variations through consistent, predictable business processes in order to deliver world-class levels of quality. The central idea behind Six Sigma is to measure how many defects there are in a process, then figure out how to eliminate them and get as close to zero defects as possible. This framework helps businesses focus on developing and delivering near-perfect products and services.

Best of Both Worlds
Each of these techniques has their advantage and limitations. Lean principles reduce cycle time more effectively, while Six Sigma is considered stronger in project management and reducing variation. Therefore, by combining the two techniques into Lean Six Sigma, we created a stronger process that utilizes the strength of each technique while compensating for their most notable weaknesses.

The following is a case study that illustrates the extraordinary results created by Lean Six Sigma. (The client’s real company name was not used; rather, I refer to it as Business as Usual Corporation “BUC”).

The Challenge
BUC is a profitable $30 million revenue manufacturing company that produces over 100 different products. As demand for its products grew, so did the number of jobs demanded of its production floor. Unfortunately, the harder the production area worked, the further behind it became --- finished goods inventories increased, but when Manchester was hired, BUC had been operating with a backorder for 13 consecutive months.

Our initial assessment revealed a number of challenges in BUC’s production process:

• Production lead time was 70 days, while actual production time was less than 10% of the lead time
• The shop was clogged with work-in-process (WIP) inventory
• Production was processed in large batches, with months of usage in each batch
• Yields ranged from 15 to 95%
• A traditional push production system based on the sales forecast was used, rather than being based on customer orders
• The Material Requirements Planning (MRP) system was utilized for scheduling, and production blindly followed the system recommendations
• Only the expeditor knew what should be worked on and when to do it
• Product flow through the production floor was disjointed
• Each part followed the same series of process steps but the time at each operation varied greatly by part
• Capacity among production areas was uneven

The Approach
Manchester was asked to help BUC eliminate the backorder situation and gain control over scheduling and coordination of production. We accepted the engagement, and through our initial assessment quickly discovered the key issues. Lean Six Sigma was the obvious choice as a framework due to BUC’s long lead times, glut of inventory clogging the system, and the yield and reliability issues which all lead to poor customer fulfillment exemplified by backorders.

Solution 1: Create Visual Communication
Our assessment revealed that although each area of the production floor was working harder, it still relied on one person in the plant to know what work needed to get done. As a result, that person had the best knowledge of production needs, but with hundreds of orders on the floor at any given time, it was impossible to keep them coordinated. We realized that some people were working on product that was not needed, instead of on product that was needed, which was part of the reason backorders continued.

In response, we designed a visual production schedule using simple magnetic dry erase boards and magnets to solve this problem. Each product line was given its own board with columns representing the sequence of operations. Each job received a magnet, which moved across the columns as each operation was completed for that job. When a job arrived in a column, the production workers completed their work on that operation, recorded the completion date on the board and moved the magnet to the next column, which in turn initiated work in the next operation.

By creating this visual schedule, every operator, supervisor and manager on the production floor could see at a glance what was happening and the status of every job in real time. Status and other information that is normally shrouded inside an MRP system became obvious. Bottlenecks were illustrated by a higher number of magnets in a column. Delayed jobs were noticeable because jobs above and below it were further on in the process. Specific known issues were marked in red and the person(s) responsible to eliminate the problem were immediately dispatched to resolve the issue. Information became quickly and widely available, much more so than with a traditional MRP system.

Solution 2: Implement Pull Scheduling
The visual schedule was a means to view and communicate across the entire production system, however it only provided half of the solution to remove the glut of work in process and backorders. The other half of the solution was placing the right jobs on the boards so that production was working on the right jobs at the right times.

BUC’s production process was driven by its MRP system, which initiated production jobs according to the sales forecast. To align BUC with the demands of its customers we advised them to initiate production jobs according to actual customer orders, not the sales forecast. As the customer orders pulled stock from finished goods inventory, replacement orders were triggered to production.

By doing so, BUC eliminated the impact of mismatches between the forecast and actual customer orders in both volume and timing. As a result, the company was no longer wasting time and inventory to produce items customers did not need and reduced backorders on items that customers had ordered. In other words, all of production was focused on building the right products. Thirteen consecutive months of backorders were ended in 22 days from implementation.

The forecast is still very useful for capacity planning and procurement.

Solution 3: Level Minimum Capacity
Our assessment revealed that some areas of the production floor were operating slower than others, so our goal was to even out the entire capacity of the production floor. The principle we followed was that systems could only produce as fast as the slowest operation in the system. Therefore, we advised management to focus their attention on the slowest area that was causing the bottleneck. BUC’s core product sells at a rate of 200 units a day, so the minimum production capacity for each area was raised to 200. The production rates for accessory products were set at 62 and 160, the rates at which customers order those products.

With this new focus on bottleneck areas, we were able to shift personnel from areas with too much capacity to the lowest capacity area, thereby increasing the flow of units through the system without adding costs. As a result, BUC increased production by 30%, with only a 10% increase in new personnel.

To keep each step in the process moving at the same rate, we continued the pull system between the production steps. This change meant that each operation only created as much work as the next operation in line could accomplish. Traditionally, in a push system, each department produces as much as it can and passes it down the line, regardless of whether or not the next operation can do anything with it. The result of “pushing” instead of “pulling” is excess WIP, longer lead times, lower flexibility, less clarity to the production line status, and more quality issues.

Solution 4: Improve Speed and Reduce Variations
Reducing the batch size was another important step in shortening the production lead-time and increasing the flexibility of production.

Traditionally, products were batched and flowed through the system in quantities of 100 or greater. Each piece in the batch was completed before the entire batch was sent on to the next production area.

We initially advised BUC to cut batch sizes in half, and as a result combined with the other changes, lead-time was reduced from 70 days to 22 days. With such good results, BUC implemented further changes and within each batch created processing batches of 5. This strategy results in each group of five being moved to the next operation when they are completed. Now, elements of each batch of 50 can be in more than one process step at any given time.

Analyzing variation in yields made additional gains in both capacity and productivity. In one product line, for example, yields varied from 15% to 80%. By utilizing a series of problem-solving tools and subsequent experiments, yields are now 45% to 80%, and additional tools and experiments indicate the potential for another 20 percentage points in improvement available in the near future with a third change proven beneficial to all yields.

Solution 5: Organize by Product Family rather than Department
Many of BUC’s products were produced in four departments (A, B, C, and D) and therefore each department worked on each product. We advised BUC to organize by product line (1, 2, 3…) with each product line having dedicated employees from departments A, B, C, and D.

Though the products went through the same basic steps, the effort needed by each department varied greatly among different parts. For example, product 1 was similar to product 2 in departments A and B, but twice and three times as long as product 2 in operation C and D respectively. When organized by department the production rate became herky-jerky, looking much like rush hour traffic, with surges of productivity as the faster parts come through, then slacks when the slow parts go through. These surges and slacks made it difficult to balance production in departments and see and measure the impact of varied process times of each product.

We advised BUC to group their products into six major product families and sort the flow of production according to these families. Parts were put into groups based on their similar production patterns. Each pattern had a maximum of a 30% variance in effort. The production process was then organized by product family, rather than by department, and production teams had representation from across functions.

Organization by product eliminated the effects of widely divergent performance in parts working through the same work area. Think of rush hour traffic. Today, cars, busses and trucks all share the same road even though they have different driving styles and capabilities. Busses make consistent stops and accelerate slowly. Rush hour cars stop only once at home and accelerate more quickly. One bus merging into traffic has a ripple effect on all the other vehicles.

In a Lean Six Sigma world, each vehicle would have its own lane, one for cars, one for busses, one for trucks and be managed separately. Bus lanes would be designed to accept frequent merging and exits with slower acceleration. The car lane can be designed for fewer merges and faster acceleration. Thereby eliminating the ripple effect that busses have on cars in traffic. Further, now that interactive effects have been eliminated, optimizing the design of the lane to the vehicle and demand is much easier.

With the move to processing batches of 5 and the production line dedicated by product family, the results were the elimination of another 10 days out of the lead-time. As a result, BUC has significantly increased speed through its system. Keep in mind: nowhere did we actually change the cycle time of each operation to achieve these results.

The Impact
By applying the principles of Lean Six Sigma, BUC was able to radically improve its productivity and responsiveness to its customers within only a nine-month period. Below is a summary of results.

The BUC example is clearly a manufacturing firm and manufacturing is an environment in which the applicability of Lean Six Sigma is easiest to visualize. However, at BUC, manufacturing represents only 40% of its total costs, and they have opportunity in the other 60% by using the tools and techniques learned on their production floor.

I’ve demonstrated that implementing the Lean Six Sigma combination can have dramatic measured results in reducing waste, improving speed and efficiency, and improving quality, which all result in improved financial results and customer responsiveness. However, there are also intangible results that all types of businesses can achieve by Lean Six Sigma that include:

• Understanding at a glance how tasks / production / projects are progressing in real time and what activities have been accomplished
• Driving improvement through a common framework, tools, and techniques
• Creating a learning environment

If you desire to achieve similar spectacular results in your business, the solution for you could also be implementing a Lean Six Sigma combination.

About the Author
Mr. Patrick Sullivan is an operations and productivity specialist with an MBA in Operations Management Sciences. His expertise in lean manufacturing, quality management, and system development and integration enable him to make effective contributions in a variety of industries. Before joining Manchester, Mr. Sullivan held operating positions in purchasing, quality, production control and executive plant management.

Working with Manchester clients in the area of productivity improvement, Mr. Sullivan focuses on improving on-time delivery and customer service, dramatically shortening lead times, lowering production costs and reducing inventories to free millions in cash. These skills impact the manufacturing floor, “back office” operations, distribution and information technology. Mr. Sullivan assists both public and privately owned companies through consulting and interim management engagements.

Mr. Sullivan’s previous experience includes process consulting with Perot Systems in Minnesota; operations and quality management at Marshall Engineering and at BTC Productions in California; and managing total quality initiatives at Environmental Care, Inc. in California.

Mr. Sullivan graduated as an Aquinas Scholar with a Bachelor’s degree in Human Resources Management from the University of St. Thomas and an MBA with a focus on Operations Management Sciences from the Carlson School of Management at the University of Minnesota and is now an adjunct faculty member at the Center for Business Excellence, University of Saint Thomas.