Taking a Value Stream Walk at Firm A
I was out walking through a company this past week, something I often do. The firm I visited had asked what I thought of their lean efforts to date and I paid a visit to find out. While flying home, it occurred to me that you might find my method and checklist of some use in your own improvement activities. So, let me share it with you.
As often happens, when I arrived at the firm the senior management wanted to start in a conference room with a lengthy overview of who they are and what they are doing to improve, focusing on their current lean "program.” After a few minutes I suggested -- as politely as I could but very firmly -- that we should delay our discussion until we had all taken a brief walk together.
I then suggested that we pick one product family and follow its value stream from the customer back to materials in receiving. Once we had selected a sample product family and started walking, I asked ten very simple questions:
1. "What are the business issues with this product?” Inadequate return on investment? Poor quality? Inability to meet customer ship dates? Inflexibility in the face of volatile markets? (If a firm doesn't know what it's business issues are, how is it going to know what to improve?)
2. "Who is responsible for the value stream for this product?” If no one is responsible for anything and everyone is responsible for everything, how can the firm improve?
3. "How are orders from the customer received?"
4. "Where is the pacemaker process, triggered by these customer orders?"
5. "How capable, available, adequate, and waste free are assembly activities?"
6. "How capable, available, adequate, and waste free are the fabrication activities feeding assembly?"
8. "How are materials supplied to the assembly and fabrication processes?"
9. "How are materials obtained from upstream suppliers?"
10. "How are employees trained in lean procedures and motivated to apply them?"
After a thirty-minute walk to answer the ten questions, I knew everything that I needed to be able to tell the senior managers just where they stood regarding their progress toward a truly lean production system. While we had looked at only one value stream, I knew from long experience that the issues we had found will be present in every other value stream. (Another walk would be required, however, to answer the parallel question of just how lean their product development system is. That would involve following a sample product design from concept to launch.)
I thought you might find the answers to these questions interesting for the real (but disguised) company I recently visited. (Scroll down to read the answers along with an action plan for the company that I derived from the answers.)
I hope that you have your own set of questions clearly in mind and that you are periodically walking your value streams to spur the achievement of leaner future states for every product family.
The Answers and an Action Plan for Firm A
This is a brief description of my recent value stream walk at Firm A, a company making a relatively high-volume, complex component for several automotive OEMs. As we walked toward the beginning of the value stream for Product A, Firm A's managers provided a quick answer to the first of my ten questions.
"What are the business issues with this product family?" Due to continuing price pressure from the two customers for the product, Firm A was losing money even though meeting a high quality standard and shipping on time. It followed that costs needed to be reduced quickly.
"Who is responsible for the value stream for this product?" This question was easy too: No one. The product (and the order) simply made its way through many departments and areas -- sales, production control, assembly, fabrication, purchasing -- as best it could with no individual assigned responsibility for managing and improving the total flow of value. At the same time, a corporate improvement group - the "Lean Team" - was making several interventions in Product A's value stream at isolated points to improve wasteful practices.
"How are orders from the customer received?" Firm A was receiving a monthly forecast and a weekly schedule from its two customers for this product family, with shipping releases controlled by physical kanban brought by milk run drivers sent by the customers. On the face of it, the use of simple kanban for shipping releases seemed "lean". But when we looked at the actual situation we discovered that kanban isn't kanban and "lean" isn't lean. One customer sent kanban every two hours with careful attention to leveling demand so that short-term production variations in the customer plant did not affect operations in the supplier plant. The other customer sent it's kanban erratically within very wide pickup windows, and a brief glance at the pattern of kanban arriving showed that this customer was actually amplifying the production variations in its own plant in its orders to its supplier. The supplier responded to these differing customer approaches in a way that was easy to see: The shipping lane for the first customer was very short, containing only the goods being assembled for the next shipment. The shipping lane for the second customer was quite lengthy (even though average demand was the same) and contained much more than was likely to be needed for the next shipment. This permitted Firm A to deal with the variations in order flow while achieving 100% on-time shipments.
"Where is the pacemaker process?" Another simple answer: There was no pacemaker. Instead, Firm A used a master schedule developed each weekend from the customers' weekly schedules and sent these schedules to each of the fabrication and assembly areas along the value stream. This was inevitably supplemented during the week by area managers re-sequencing orders to deal with changes in demand and with production problems along the value stream. This was not at all lean: No takt image and no ability to know within a few minutes whether operations were supporting the customer.
"How capable, available, adequate, and waste free are assembly activities?" A recent kaizen at Firm A had created an assembly cell combining a number of assembly and sub-assembly activities formerly conducted in different areas of the facility. The processing steps had been placed in close proximity in a U-shaped area and the area manager for assembly stated that Firm A had now achieved continuous flow assembly. However, only a moment's observation showed that work was poorly balanced in the cell, with little evidence of truly standardized work, and that small piles of inventory were building up between each step. In addition, the Production Analysis Board next to the cell showed clearly that output was varying markedly from hour to hour. The explanations in the margin of the Board showed that the processing machinery was both capable and reliable but that materials shortages often stopped the cell. My eyes told me immediately that the cell should be able to run steadily at its planned output, based on takt time, with about half the operator effort. This should have a major effect on costs.
"How capable, available, adequate, and waste free are fabrication activities?" A recent kaizen led by Firm A's Lean Team, had also created two fabrication cells for the product, with the first cell directly feeding the second cell so that they were effectively linked as one cell. However, a moment's observation and a look at the Production Analysis Board for both cells showed major problems with capability and availability. Indeed, the cells together seemed to be stopping - either due to producing defective parts or the inability to cycle at all - about 20 minutes out of each hour. As a consequence, a large amount of overtime was being run and considerable buffers of work-in-process were kept after the first cell and at the downstream end of the second cell. Clearly there was a need for the Lean Team to focus immediately on both quality and maintenance if costs were to be reduced.
"How are orders transmitted up the value stream from the pacemaker process?" A moment's observation of the area managers in assembly and fabrication showed that a key element of their jobs was to continually adjust the schedule to deal with demand shifts downstream and process problems upstream. What was needed instead was a simple supermarket system between each step and a simple pull system to trigger work by the upstream process only as parts were needed by the downstream process. Doing this would reduce the total amount of inventories needed and free up management attention for further improvements in the value stream.
"How are materials supplied to the final assembly and fabrication processes?" The production control and Logistics manager proudly showed off the new "waterspider" system of supplying materials to the fabrication and assembly areas from a receiving supermarket. The waterspider circulated through the plant once an hour to deliver needed materials to each production area and to collect finished goods for transport to the shipping area. What could be leaner? Actually, everything. The waterspider was not involved in distributing production instructions and so had only a vague idea of what each production area would need. The simple solution was to put an ample supply of practically every part number on the lengthy tugger train so that whatever part was needed could be supplied. A moving warehouse! In addition, there was considerable confusion in the storage locations for each part number and no Plan for Every Part showing exactly how it would be reordered, packed, shipped, received, placed in the supermarket, and distributed. While some parts expediting might have been eliminated by the new materials delivery system, it was achieving only a fraction of its potential benefits.
"How are materials obtained from upstream suppliers?" Supplier shipments were triggered by Firm A's master schedule, which was itself being adjusted from hour to hour. As a result, the area manager in receiving was continuously working with area managers on the floor and with purchasing to change orders to suppliers and keep production running. This manager seemed to be very proficient at this task, but why was it necessary? Couldn't suppliers instead be put on a simple pull system with appropriate leveling so that any short-term variations in the plant's performance would not be inflicted on the suppliers? And couldn't supplies be collected by frequent milk runs in small amounts rather than by the current direct ships every few days in large amounts?
"How are employees trained and motivated in lean procedures?" This was perhaps the most shocking aspect of Firm A's operations. Most of the production associates were actually employees of a manpower firm and were working on short-term contracts. This held wages down and discouraged recent efforts by several unions to organize the plant. But it also meant that standard work was hard to maintain, multi-skilling was difficult to implement, and that no production associate could reasonably be expected to contribute to kaizen activities. From observing the efforts of production associates, I concluded that the savings in cost per labor hour were very likely more than offset by poor productivity during each hour.
These questions only apply to physical production and I could have taken a similar walk through product development or the extended supply chain. However, the point for current purposes is simply that a walk taking only thirty minutes was quite sufficient to determine just how "lean" Firm A is and to come up with a prioritized list of steps the firm should take.
As for "lean", my simple sum-up was "just barely on your way and without a clear plan." My proposed action plan is as follows:
1. Clearly identify all your value streams and clearly state the business issues confronting each. For the specific value stream in question, set a cost reduction target that will produce an adequate return.
2. Appoint a value stream manager for each product family to both manage and improve it, addressing the business as well as operational issues.
3. Work with customers to smooth demand and eliminate amplification. (And, at a minimum, use finished goods as a buffer to smooth the flow of products along this value stream.)
4. Send production instructions up the value stream by means of simple pull loops with leveling from the pacemaker process at the final assembly cell.
5. Make the assembly and fabrication cells into real cells by tackling capability, availability, and work force utilization issues.
6. Establish a paced withdrawal system for materials with short intervals (perhaps twenty minutes) and a rigorous Plan for Every Part.
7. Work with suppliers to smoothly transmit demand to the next firms upstream and to get frequent deliveries on a precise schedule in small amounts.
This list is only the beginning for Firm A, of course, but it is a real beginning leading toward a true lean business system instead of another "program" involving isolated interventions with doubtful results.
|Gemba Walks, 2nd Edition|
|Seeing the Whole Value Stream, 2nd Ed.|
|Lean Thinking, 2nd Edition|
|The Machine That Changed the World|
- Learing to See the Whole Value Stream: The Power of Value-Stream Mapping
- Sustaining Lean Goals by Taking a (Gemba) Walk
- Forward to Fundamentals
- Managing to Learn: Part 1 - How Lean Leaders Create Productive Problem-Solvers
- The Power of Purpose, Process, and People
- Lean Management & the Role of Lean Leadership
- Lean Solutions