注册 登录  
 加关注
   显示下一条  |  关闭
温馨提示!由于新浪微博认证机制调整,您的新浪微博帐号绑定已过期,请重新绑定!立即重新绑定新浪微博》  |  关闭

论老子

道,领导也。领导必需要不断呼唤,教导下属以及以身作则。下属的过和错皆因领导懒惰。

 
 
 

日志

 
 

Chapter 24: Quality Losses  

2012-06-24 11:45:00|  分类: Buffer Mentality |  标签: |举报 |字号 订阅

  下载LOFTER 我的照片书  |

John said, “Quality losses are tangible losses. Everybody in the manufacturing industry knows about this fact. Why make you think that quality losses had to do with buffer mentality? Or for that matter, how buffer mentality leads to more quality losses?”

I said, “This got to go back to the philosophy behind the Toyota Production System. It says, ‘surface problems for immediate action. Any quality issue must be surfaced for immediate resolution. Otherwise, the factory will continue to suffer quality losses. Do not cover up any problem and bury it out of sight’.

This is the reason behind why Toyota Motors Corporation earns a bigger margin of at least US$500 more than an equivalent compatible car model made by any other Japanese car company. ”

“But it is quite difficult to imagine why Toyota Motor Corporation eliminates quality losses much better than the rest of the automobile companies. Neither it is easy for me to relate to buffer mentality as the chief reason why most other companies could not do it better than Toyota Motor Corporation,” queried John.

I asked, “Shall I make use of the competition among the mobile phone companies to illustrate the importance of buffer mentality in the prevention of many of the mobile phone makers in China from aggressively pushing their quality level better than their competitors. In fact many of these mobile phone companies are on the brink of fighting for their survival?”

John replied, “Okay, this is an interesting industry where there are almost a thousand companies in Shenzhen alone competing to produce imitations or close clones of famous brands such as: Nokia, Samsung, Motorola, Sony Ericson mobile phone sets.

Two years ago, the profit margin of a clone mobile phone is around 150 Renminbi. A year ago, the profit margin had gone down to 60 Renminbi per mobile phone set. Today, in 2008 the profit margin had dropped down to 20 Renminbi per mobile phone set.

The question is how much lower the profit margin can go before many of them go bust.

Of course, the single determining factor in preventing the profit margin from going under water is to ensure the product quality is good. A significant yield loss will definitely eat into their profits and possibly causing them to make a loss on every unit they produced.”

John said, “Agree.”

 

Illustration #1: Cleaner environment  

 

I explained, “I visited GEMS a mobile phone set maker in Shenzhen in October 2006. The general manager was inviting me to assist him to reduce his cost of production for mobile phone sets.

He explained to me, ‘As late as one year ago, we were enjoying very good profit margin. Our biggest problem was the yield loss at the paint shop. It stood at slightly above 8%.

Yield losses had been a major factor that eats into our profit margin. If we bring down the yield loss, every single percentage savings contributes directly to our net profit margin. For this reason, the original equipment manufacturers pay a lot of attention on our quality performance. To them a vendor who can reduce its quality losses is in a better position to reduce cost per unit and thus, will be a more reliable supplier.   

Two years ago, we suffered yield loss of around 11%. We designed an innovative spray paint production line and we managed to bring the yield loss down to 8%. This is the best in the industry currently among the Chinese mobile phone manufacturers. The Japanese mobile phone makers are the pace setter in term of quality and possibly performing a shade better than 8% yield loss. We do not think anybody can do better than this quality level.’

In my mind, if he said he is already matching the Japanese – the best in the industry in term of quality or specifically yield loss, he will not push his people to continuously reduce yield losses. This is a self-fulfilling prophecy.

To believe that the industry is at best equal to them in term of yield loss does not mean that some of the competitors can reduce the yield loss further. To me, this certainly is buffer mentality.

When I was at their plant, I had to put on a smock and a cap before I can walk into the spray paint shop. This must be a class 1,000 clean room area. I walked down the spray paint production line and I notice it is crowded with production operators. I asked him, ‘Why are there so many operators deployed in this confined space?’

He replied, ‘This production line is a high speed line. In order to load and unload the phone bodies onto our specially designed jigs that travel along the conveyor belt, I need to deploy a large number of operators.’

I asked a prodding question, ‘What is the main cause of the paint defects?’

He replied, ‘When a dust particle lands onto the surface of the phone body, a subsequent spraying of paint on top of it results in a tiny rough surface. This is an out-right reject.’

I asked, ‘Do you think in a clean room environment, it is important to have the least amount of movement. This is because human movement will kick up the dust into the air and the dust particles therefore will fall onto the conveyor line?’

He replied, ‘Yes, I do. But it can’t be helped. I need to watch my productivity. I cannot let the conveyor line slows down intermittently because of slow loading of non-spray parts onto the conveyor belt and subsequent unloading of the sprayed parts.’

I explained, ‘Productivity is often the excuse for not pushing for better quality. It is a common excuse. Any excuse to me is buffer mentality. It prevents you from looking hard at how to reduce yield losses. I can help you to reduce the number of manpower deployed in the spray paint clean room area by two-thirds. Would you think the quality will improve by a significant level?’

He replied, ‘Yes, that will definitely give me a percentage point better in quality. Moreover, the reduction in labor is substantial. I buy into your idea. But how do you reduce the manpower requirement?’

I replied, ‘Of course, I have the know-how. I did it many times over in more than ten companies.’

In my mind, I was not ready to tell him my solution. I want to diagnose the other quality issues that he was facing. I continue to ask him another question, ‘How frequently do you carry out maintenance for the spray paint conveyor line?’

He replied, ‘Once in a month. The preventive maintenance exercise takes more than a day. So it is a significant loss in utilization capacity.’

I asked, ‘By the way why do you carry out preventive maintenance?’

He replied, ‘Along the conveyor line, the pallet keeps knocking against one another as they travel along the entire production line. This cause dust particles to be generated and flies into the air around the spray paint section.

Even though we apply a constant stream of downward blowing air draft, dust still settles on the surface of the phone body causing a rough spot. We know this is almost impossible to eliminate. That is why we settle for a historical best of 8% yield loss. This is already the best quality level among the mobile phone making industry.’

I suggested, ‘Do you think it will reduce the yield loss significantly it you were to increase the frequency of preventive maintenance to once a week?’

He replied, ‘Yes, it may increase the quality by another percentage point. But the capacity utilization may have to suffer again.’

I asked, ‘Don’t you think you are using utilization capacity as an excuse again?’

‘You are right,’ he said, ‘’But how not to pay attention to capacity utilization? It is crucial to bring down the cost of production.’

I did not reply to his question. I peeped into the spray chamber and I noticed the spray chamber is a very small confined space where there is a spray gun and an air gun place side by side. But outside the spray chamber, the mobile phone bodies though already placed onto the fixtures that sit on top of the pallets, moving around the conveyor belt of the spray shop are exposed to the environment. Though this is already a class 1,000 clean room, there are still lots of dust particles being kicked up into the air.

I asked, ‘Do you think it is important to build a box to cover up the entire length of the conveyor belt? In this way, you can prevent the dust particles that were kicked up into the air from landing onto the phone body.’

He replied, ‘That is a good idea. But it adds costs and the downtime for preventive maintenance will be much longer.’

I smiled to him, ‘Well, I have given to you three good solutions to increase your yield. Whether you want to take up any of these solutions depends on your buffer mentality. Or just any one of the excuses you had mentioned to me earlier is sufficient to knock you into status quo and live with the 8% yield losses forever.’

He nodded his head.

John, sometimes, the solutions to eliminating a quality issue or reducing yield losses is pretty straight forward. But what makes you stop applying the solution? That is buffer mentality. Don’t you agree with me?”

John nodded his head, “Yes, many of us always try to look for the seemingly impossible holy grail to a quality problem when the solutions could be so simple as controlling dust particles, more frequent schedule for preventive maintenance and simply enjoying a clean environment.”

I said, “John, in the next example, I will tell you how removing the mental block that inspection is a waste and therefore, is not a value-adding activity resulted in zero defects quality level that no one ever believes it can happen when the first pass yield was then only 60%.”

John expressed his opinion, “Yes, inspection is a non-value adding activity. That is how everybody classifies inspection activity. The industrial engineers are the ones who will argue strongly against classifying inspection as necessary. Almost every production manager does not agree that their production operators should carry out any inspection activity. That is the quality control inspectors’ job.”

“In the next example, I show you the reverse. By adding in inspection as one of the production operators’ sub-tasks, the productivity level shot up by 10% and the quality level hit 163 consecutive days of zero defects. This is amazing, isn’t it?” I asked.

“It is conflicting. For all I know, quality and productivity is an either or kind of situation. Bette quality means lower productivity. Or higher productivity means lower quality. Not both up at the same time. A real-life case where zero defect quality was achieved and at the same time, yet raised the productivity by 10%? That is unheard of,” said John. 

 

Illustration #2: Quality must be built-in

 

I explained, “When I first walked through the production line in Archive Singapore, I observed that the production workers sat on both sides of a conveyor belt which was moving at a constant speed. When the first worker finished a small task of assembling the chassis of the tape drive she placed it onto the conveyor belt which then moved it down to the next worker. The second worker picked it up and assembled other parts onto the chassis and put it back onto the conveyor belt. As the conveyor belt moved the chassis from one worker to the next, the last worker finished the assembly and placed the completed tape drive into a tote box.

I asked the industrial engineer if all the different tasks carried out by the workers were equal to the speed of the conveyor belt. She replied, ‘No. However, the speed of the conveyor is tuned to the speed of the worker who does the slowest task.’

‘What was the output rate expected from the assembly line?’ I asked.

The industrial engineer replied, ‘The conveyor speed was set at 0.1 meter per second. The space between two adjacent workers is one meter and its takes exactly 10 seconds to move one unit of the product from one worker to another. That works out to 360 units an hour.’

Knowing that a worker could not work at a consistent pace throughout the day I asked about the number of units they were producing an hour at present.

The industrial engineer replied, ‘We produced 324 units an hour on average. That is about 90% in labor efficiency.’

Well. In my mind, that is a productivity level of 90%, a considerably wide margin of 10% away from 100% productivity measured against the slowest worker. This means all the other workers were working at much lower than the 90% productivity level.

However, the most critical problem was not the productivity level. The first pass yield of the assembly line was merely 60%. In other words, 40% of the outputs went into the ‘pile of bones’ waiting for its turn to be reworked.  

I implemented two strategies to solve both the productivity and quality problems faced by this assembly line.

First, I stopped the conveyor and introduced Kanban trays that were designed to hold only two units of the tape drive chassis. Between two workers, I placed one Kanban tray allowing a maximum of two units of the chassis as Work-In-Progress between them.

Next, I introduced a simple rule - If the outgoing Kanban tray is full, the worker shall stop producing. Otherwise, she should continue producing. The workers know that if the outgoing tray is empty, the downstream worker would be idle waiting for her to pass down the next chassis. The upstream operator knows she has to speed up. The workers know how to read the situation when to speed up and when they can slow down to a normal pace.

The productivity went up to more than 95% in just one week.

Second, I introduced a quality concept that very few companies put into practice because the division of labor had split the roles of a production worker and the quality control inspectors. Most companies are organized in clear departmental roles. The production manager told his workers to be responsible to produce goods and meet the output target and not to inspect the product for quality. The assurance of product quality rests with the quality control inspectors. This is where the root cause of the problem is.

Imagine. When the whole bunch of production workers didn’t think it was their responsibility to produce quality products, there was no way the quality control inspector could inspect quality into the product.

To make things worse the industrial engineer could not agree with me initially that the workers should inspect the quality of the product they produced. Inspection is not a value-adding process and everybody knows that. Therefore, to the industrial engineer, inspection should not be factored into the task of a production worker. That will lower the production output rate.


Figure 24-1: A Kanban system aims to reduce the number of units at risk



Chapter 24: Quality Losses - 浪里行舟 - 论老子
 
 

 

 

 

 

 



I managed to convince both the production manager and the industrial engineer to allocate time for the workers to inspect the quality of their work. Of course, this was accompanied by a few other programs that I introduced in parallel.

In less than one month, the productivity approached 100% every day and the first pass yield went up to 99.7%. Factoring in the inspection should have lowered the production rate but why is it much higher now?

Everyone was puzzled. I was not.

I observed sometimes the downstream workers returned the chassis to the upstream worker to get it reworked. After getting the workers to inspect their work, the passing of rework unit back to the upstream worker came to a complete halt. This accounted for the productivity approaching the 100% productivity level everyday.

Incidentally, asking the workers to inspect their own work actually means telling the workers to build quality into the products. If they find anything wrong with their workmanship, they immediately rework it. They do not want the downstream worker to pass a poor quality unit back to them for rework. That is, the essence of built-in quality.

 

Figure 24-2: Immediate feedback system



Chapter 24: Quality Losses - 浪里行舟 - 论老子
 
 

 

 

 

 

 

 

Months later I began to notice that on some days the production line did not produce a single piece of reject. That struck something in me.

In 1980 when I started out in my first job with Hitachi Semiconductor Malaysia, I saw a neon-lit signboard placed right in the center of the factory compound. It read, ‘Zero Defects’.

Back then I could not understand how a factory can achieve zero defects. I thought to myself, “Ah! It is another publicity drive by the company to create the awareness for quality”.

But right in my face, the production line had indeed begun to produce goods with zero defects. Zero defects now make sense to me.

While traveling to work, I often see big notice boards erected in front of most construction sites. The first line reads, ‘Number of days accident-free – xxx’ I told myself, why don’t I do something similar? Instead of ‘number of days accident-free’, I replace it with ‘number of consecutive days of zero defects’.

In one of the weekly JIT[1] review meeting I told the managing director and his staff that it was about time to introduce a campaign for zero defects. Every one of them laughed at me. Fresh in their memories, the production lines were throwing out 40 percent rejects one year ago. Now I was asking for zero defects. They thought I was joking.

I continued to push my idea to them, ‘Well! Give me a try. Anyway it is not going to do any harm to the company. Getting the workers to drive towards zero defects is certainly the best thing you can do to create quality awareness.’

I bought a few pieces of white boards. In each of them I wrote two lines: ‘Current consecutive number of days zero defects ____’ and ‘Record number of days zero defects _____’. I asked the production supervisors to update the board in the morning every day.

All the production supervisors looked at me in disbelief. In their mind they were thinking, ‘Are you nuts? There is no way we can achieve zero defects.’

Even though in total disbelief, the supervisors updated the zero defects communication boards diligently. It all began with one of the production lines producing zero defects for one day. ‘Ah! That was by chance,’ exclaimed the production supervisors.

But that number slowly crept up day by day. When a defect cropped up, the production supervisor set the count back to zero and started counting the current consecutive number of day zero defects from one. But down on the second line, the record number of days zero defects had become a positive number.

The managing director and his staff conduct MBWA – ‘Managing By Walking Around’ every Friday and they began to notice the incremental rise in the record number of days of zero defects.

The managing director commented, ‘Hey! Eric means business! He was so confident of the ability of the production people before he proposed to me to launch the zero defects campaign. We almost failed him.’

One of the production lines managed to achieve a record of 163 consecutive days zero defects. That is about half a year. Unbelievable achievement! But it happened.”

John exclaimed, “That is really marvelous. Not only that you eliminated the buffer mentality among the industrial engineer and the production manager, everyone in the factory from the CEO down to all the operators realized that quality losses is not a thing that is a necessary evil. It can be reduced to zero quality loss.”

I summarized, “Frankly speaking, zero defects can only happen in a Just-In-Time production environment. To be more specific, the production system must be running at a one-piece-flow system. A lean production system is not good enough because most lean production systems are short of achieving the true one-piece-flow production system.” 

John nodded his head. He lamented, “In the lean expert workshop, the facilitator teaches us that the one-piece-flow system guarantees the shortest lead time. Now you have enlightened me that the reason why you always implement the one-piece-flow production system ahead of other lean techniques - It is quality.

Without assuring a close to zero defects environment, it is impossible to eliminate buffer mentality among the whole bunch of people who work on the shop floor or the supporting functions. ”    

In admiration of his sharp mindedness, I said, “John, you are a fast learner. You are worthy to be my student.”



[1] JIT means Just-In-Time. JIT production system is the name used to describe the Toyota Production System back in the late 1970’s before the currently made popular term, Lean Production System. However, they are not the same thing. Please refer to the Introduction.

  评论这张
 
阅读(188)| 评论(0)
推荐

历史上的今天

在LOFTER的更多文章

评论

<#--最新日志,群博日志--> <#--推荐日志--> <#--引用记录--> <#--博主推荐--> <#--随机阅读--> <#--首页推荐--> <#--历史上的今天--> <#--被推荐日志--> <#--上一篇,下一篇--> <#-- 热度 --> <#-- 网易新闻广告 --> <#--右边模块结构--> <#--评论模块结构--> <#--引用模块结构--> <#--博主发起的投票-->
 
 
 
 
 
 
 
 
 
 
 
 
 
 

页脚

网易公司版权所有 ©1997-2017