Return to Article List
 

How Lean Transformation and Simple Technologies Are Accelerating the Supply Chain

by ASCET | December 14, 2005 |

Solectron's Marc Onetto explains how lean transformation and simple technologies are accelerating the supply chain.

ASCET: Solectron has implemented lean logistics extensively. Have you achieved a pull model of demand-centric inventory with this approach?

MO: Lean logistics is a journey, not a destination. If you asked Toyota today if they have perfect demand-centric inventory, I think they would tell you that they still have things to improve – 57 years after they began. I’ll be very modest and tell you that we are far from having achieved perfect demandcentric inventory. We have implemented a lean transformation, which is much more than just lean logistics. In fact, in order to have some form of a pull model there are many other steps that need to be taken before there is the maturity to function on pull. At Solectron, we have first concentrated on totally transforming how we manufacture internally and how we pull between our own plants. We have also progressed toward increased collaboration with suppliers and customers, but is it prevalent all across the company? Far from it.

ASCET: What needs to happen to make collaboration a supply chain reality?

MO: With our suppliers, I would say we are at the beginning of the journey. We now have perhaps 5 percent of our suppliers on perfect pull. We have a long way to go; our objective is to have about 50 percent of our suppliers on pull within three years. With our customers, we are still very much on a forecast basis, and we are working very hard to transform to pull with some key forward-thinking customers.

The forecast model generates an incredible amount of waste for both  sides it’s bad for us; it’s bad for our suppliers; but it’s also very bad for our  customers. The old model was to secure the supply chain through forecasting, which was a totally hopeless task. The paradox of the supply chain  industry is that innovative products are introduced to the market every six  months and the supply chain takes six months to deliver a product. Securing  the supply chain through forecasting is a totally hopeless task in an innovative industry, because by the definition of innovation you don’t really know  what the customer will like in six months’ time. We are fairly convinced that  the only way we can deal with this is by accelerating the supply chain, and  the way we do that is by passing the demand information as fast as possible. That’s why collaboration becomes a key element.

People have paid for the old model of ‘it’s not my problem; it’s the other  guy’s problem.’ Passing the problems on caused a lot of regret, and it was an  expensive lesson for people to learn. Now there is a much more open attitude to collaboration. We are very convinced that this is the only way we can  work, both with our customers and our suppliers.

ASCET: Is Solectron accelerating the supply chain primarily through faster information sharing with collaborators?

MO: Yes, but the information has to be useful. There are people who have tried to accelerate forecast sharing. What they do is accelerate the movement of useless information. The key information is, what does the end customer want? That is the information that has to be accelerated.

ASCET: Are you using different metrics and techniques to collect information?

MO: Yes. For example, one of the metrics that we have put in place is the speed by which we can configure an order that we’re receiving. The order is expressed in some form of marketing functions, and we have to transform that functionality into hardware. The metric is, how fast can we configure the order?

Another interesting metric is how many orders will be physically configurable. For example, you receive an order for a car, and you realize that they want six wheels, which is of course not available. With one key customer, we were able to eliminate about 20 percent of such orders. The orders were sent back immediately because they were physically impossible to make. We absolutely committed to figuring out what an order means and configuring it in less than five hours. This commitment forced us to do a lot of investment work along the way.

Another metric that is very important to us is a metric called delivery to customer request. Once we accept that the No. 1 driver is the end customer order, then we have to be able to deliver to that customer request. Once we have a time frame – for example, we have a customer order that we are to deliver in five, 10 or 15 days – we can commit to delivery. When we get the demand we’re going to run like mad to make that demand, and we can deliver 99.5 percent of the time.

 

ASCET: Does Solectron’s data warehousing offer real-time data visibility to your vendors?

MO: This issue of visibility is a double-edged weapon. Many people thought that we would resolve everything by having great visibility into everybody else’s  factory. But visibility is not of the essence. What is of the essence is the process by which we pull on each other and the fact that vendors have a very predictable way of producing, which is totally synchronized with Solectron plants.

Visibility is like saying, I can see what is happening, so I can improve the process. But visibility won’t help if a machine has a four-hour changeover time. There’s nothing you can do with visibility to make the changeover happen faster. So there is a fallacy in thinking that if I could only see what my partner has, I can fix it. What matters is a high level of discipline. Once you have a disciplined, predictable supplier, then you can synchronize your production with his production. Which means that he’s on pull with you. When you stop, he stops; when you accelerate, he accelerates. In Japan, I’ve seen some cases where people will fill supply plants every 50 minutes. So the plant of the supplier is totally synchronized with the plant of the customer. Then there are other cases where, because his cycle time is different from yours and his changeover time is different from yours, there you’re going to need a little more inventory between you in order to synchronize  the two cycles. So what matters is that we’re synchronizing the way we work; that is, that we are passing the right signal at the right time.

ASCET: What technology is used to achieve that level of synchronization?

MO: The technology originally used by the Japanese creator of the system, Toyota, was called a kanban, which is a plastic card. Originally the card was made of cardboard. Up to now that has been the best technology. When the supplier plant is 5,000 miles away rather than five miles away, you replace the piece of cardboard with a computer signal based on the same logic as the kanban card, which is that a signal is emitted when you have used a certain number of parts and the kanban card represents the minimum of the quantity; i.e., the economy quantity needed. That’s all you need.

Now, what we have done in the Western world is create very complex technology that has turned out to  be totally useless when we use it incorrectly. For example, EDI (electronic data interchange) is a beautiful technology if you emit the signal based on usage. If you use EDI to transmit a forecast, which changes every week by 300 percent, you create a maelstrom in the supply chain, and that brings no value.

Within Toyota, the concept of kanban – which by the way does not mean “inventory” but means “signal” – is a signal to your supplier that you have used a certain quantity of parts and you need to be refilled. The card, the kanban, is that physical signal. The card also represents how many units there are in that refill. If you look at, for example, Toyota windscreens, their kanban is about 20 windscreens, so the kanban is set to send a signal to order 20 more windscreens. When the supplier is five miles away, you don’t even need a computer system. Now if your supplier is 2,000 miles away, you bar code the kanban card, then that becomes the computer entry. The computer is only a production recoding system; it has no intelligence. The intelligence is automatically generated by the way your kanban is set up. All you are doing is sending an electronic signal, usually known as an EDI. It’s the same logic as the little piece of cardboard.

The computer is only a production recording system. The EDI tells you that the first zone is the quantity, the second zone is the description, the third zone is the price and the fourth zone is the custom code, etc.

That is EDI well done. However, many people have used EDI to transmit a useless signal, such as forecasts. What they do, they’re very proud to tell you, is every week they make five forecasts. So every week they send hundreds of thousands of signals to suppliers for the next six months, which forecast the demand for the next six months, and they change that every week; this creates a totally useless maelstrom. It makes a fortune for the people who transfer information. Those people love it. Everyone else should do something else.

ASCET: Are you seeing more of a trend toward supply chain industry consolidation?

MO: We are now in a rebuilding phase of the industry, and I therefore think that the people who have survived the bad years are now perhaps in better shape to serve  customers. The key players are now well-established, and I personally don’t see any major consolidation in our industry. Now these kinds of statements can always be taken with a grain of salt. I do believe that there will be consolidation certainly between what we call the tier 1 and 2 and the tier 3 industries. That is, the small  players will have more and more difficulty existing in an economy that is increasingly global. At the tier 1 level, for Solectron and our major competitors, I don’t see significant consolidation.

ASCET: Are you seeing more of a response to a growth economy in the industry, and does that change the way that you are thinking about supply chain?

MO: I see two trends. The first trend is there will be increased outsourcing. Some industries, such as the computer and telecommunications industries, have done quite a lot of outsourcing. However, some industries will need to do more outsourcing of electronics because manufacturing electronics is a specialty. Industries such as industrial, medical and automotive have more and more electronic components in their products, and they will increasingly require our services. Five years ago, 10 percent of the value out of a car was electronics, and today it’s more than 25 percent. Moreover, most of the quality issues in a modern car are electronics issues, not mechanical issues. Which is another indicator that perhaps car manufacturers should think about leaving the electronics manufacturing to people like us who do that for a living.

The second trend is a continuation of growth. This trend is very strong because electronics is penetrating many more sectors. We are in a positive phase. We had very good growth in 2003; in 2004 it slowed down, but overall the trend is very good.

ASCET: What about an environmentally friendly supply chain? Does the existing technology and industry leadership support a greener supply chain process?

MO: To the first part of the question, absolutely yes. There is a very strong movement in the world right now for what  we call in our industry “reduction of hazardous substances,” also known under the generic name of “leadfree.” What this means is the elimination of substances used in construction of electronics products that can be hazardous to human beings in the long term. There are many more substances besides lead. There is a lot of normal pressure from green politics, which is very good, but there are also regulations. For example, the European Union is in the process of enforcing strict industry regulation regarding lead-free or lead reduction. Our customers need to comply, and some of them will even do more than comply; they will try to be greener than required because they believe in it or because they think it’s good for their image. We are working very hard to support our customers in this area. Our work is a combination of scientific research,  processing work and redesign work. But there is also a strong dimension of green specific to what we are doing at Solectron, which is the green of lean. Lean logistics, or the Toyota production system, is centered around the concept of eliminating waste, of not tolerating defects.

If you go to a Toyota plant you will see that the plant stops as soon as somebody spots a defect in a car. Until the root cause of the defect is found and resolved, production does not restart. This is called jidoka in Japanese, and it literally means to stop a manufacturing process when something goes wrong. By stopping production to find the problem, waste is reduced over the long term, because if you produce defective products, it goes in a bin, which means it pollutes.

Lean transformation is also a great contribution to the overall aspiration of the world to pay attention to and respect the environment. For example, one of the metrics we use is the size of the boards that we scrap. That’s a major metric in lean, and we have reduced scrap at Solectron by factors of five and even 10 in  some cases. So that’s 10 times fewer things to be disposed of. We believe very much in an environmentally friendly supply chain, and we work on this both in terms of making scientific progress and also in terms of eliminating the waste that is generated in the supply chain.

ASCET: What emerging technologies or ideas are interesting to you right now?

MO: In terms of technologies, it’s a process technology – the implementation of Six Sigma statistical technology to predicting the behavior of products both in terms of manufacturability but also reliability. There is also very exciting work going on in collaboration with our customers in terms of technologies like RFID, Bluetooth  and WiFi. Now, in applying some of that technology into the logistics of the supply chain, it is only useful if it is serving lean discipline. Some of these technologies, such as bar coding, are very useful because they allow us to transmit a very clear signal to a partner. RFID can be very useful in allowing us to know immediately what to pull from suppliers or to deliver to customers. The only real, useful information that needs to be transmitted as fast as possible is the actual demand from the end customers. How many of these new cell phones are being ordered today? How many of the new iPods or MP3 players are being sold today? That is the big deal.

For us at Solectron we have a clear vision. We want to be the people who resolve the electronics industry paradox of fast product development coupled with a slow supply chain. We want to be the people who are going to bring a faster supply chain solution to our customers. We have a very simple vision, a very modest vision. We want to be the Toyota of our industry. That’s who we are and what we will be doing for many years to come.

 
Return to Article List