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 9: Misused of percentage  

2012-06-24 12:26:54|  分类: Buffer Mentality |  标签: |举报 |字号 订阅

  下载LOFTER 我的照片书  |

“This is an easy one for you,” I said, “On Feb 17, 2008; I went for a plant tour to Honeywell Turbo factory in Shanghai, China. This site is already in its phase 4 implementation of the HOS-SIF (abbreviation for Honeywell Operating System - Standardized Implementation Framework). The plant was excellent. Most of the performance indicators were above the goals.

However I noticed the quality indicator was measured in percentage. The chart showed that the yield loss has been hovering from 0.8% to 1%.

I asked the HOS site leader, ‘What does your management think of the quality level?”

He replied, ‘Excellent.’

I pursued, ‘Does your management pay much attention to continuously bring down the yield loss. What is this year goal for yield loss?’

Actually, I had already seen the goal line across the chart has not been changed for more than a year.

He replied, ‘No, not much. We kept the goal line at last year level.’

I suggested, ‘Will your management want to convert the yield loss from a percentage unit of measure to a part per million (abbreviation, ppm) unit of measure?’

He asked, ‘If there any difference?’

I explained, ‘1% is equivalent to 10,000 ppm. 0.8% is equivalent to 8,000 ppm. At 8,000 ppm, the yield loss can be considered a little too high. There is room for improvement to bring it down to say, less than 5,000 ppm.’

He concurred, “I see your point. It magnifies the magnitude of yield loss.’

Of course, I did not comment anything about the use of percentage point is an indicator that tells me that the management prefers percentage over the ppm unit of measure out of buffer mentality. That would be rude.”      

“Yes, this is a straight forward example. I have seen both percentage and ppm being used in several companies. But I did not see it as buffer mentality. You did,” said John.

“Let me illustrate the buffer mentality in the use of percentage. This could be considered as a classic case study,” I said.

“In the inkjet division of Hewlett-Packard in Singapore I audited the management function of the supply chain department. Knowing the fact that Hewlett-Packard was renowned for being the best customer among all its vendors, I really wanted to see what makes Hewlett-Packard procurement function ticks.

I already knew that Hewlett-Packard always pay on time but that is not before a two-month wait after the invoice has been submitted to HP finance department. This is mere tardiness in honoring payments and I don’t think this alone suffices to make Hewlett-Packard the best customer. There must be something more fundamental to this obvious fact.

I asked for the management objectives set for the procurement department. One of them reads, ‘Percentage of non-conformance materials supplied by the vendors and the goal set for it was at 0.25%. The monthly report only showed the achievement for the month of November 2005 and it was at 0.19%. That was 0.06% point much better than the goal. Certainly it was considered a great achievement. No question about that?

You are wrong.

 Figure 9-1: Presentation of November 2005 non-conformance



      I suspected there are two underlying short-comings with this way of presenting the monthly performance by the procurement department.

One, it did not show a trend over the last twelve months. Why am I asking for a trend chart? This is because a twelve-month trend chart will depict how well the procurement department has been doing over this longer period of time. Is it trending up or trending down or maintaining at about a consistently flat level?

The twelve-month trend chart certainly will give us an answer whether the situation is deteriorating or improving or just hovering about the same level without any change in its performance. Of course, management wants to see an improving trend.

However, by a simple stroke of presenting the statistics for the month of November only, the reviewing manager was effectively cut off from reading the bigger picture of how its performance has been doing over a longer period of time. He cannot read whether the situation is deteriorating, improving or flat. This is a fine act of burying the reviewing manager’s head into the sand like an ostrich.

Two, was the goal set too lenient?

If the picture of a twelve-month data was plotted as a line chart, the reviewing manager would at least able to make an objective reference of the goal versus the performance of the longer twelve-month period. In other words, he can assess whether the goal set was too lenient or not.

Of course, if right from the first month, the data was already beating the goal, obviously the goal was set way too lenient. If all the twelve points plotted on the chart hovered below the goal line, it become clearly a situation where either the goal was set too high or the team did not do enough to try to achieve the goal.

If the twelve points were to dance above and below the goal line on quite an even spread of points, the goal could be assumed to have been set at about the right level or simply put, the team had not done enough to beat the goal.

 Figure 9-2: A trend chart tells the performance over a period of time

 Yes, a picture speaks more than a thousand words. Anyone who doesn’t know much about the functions and responsibilities of the procurement department can read a well-presented graph or chart and makes a meaningful and correct judgment on the performance of the procurement department.

As I was a consultant trained with a pair of very sharp eyes for details I looked out for more than this. I query the fundamental question of why this objective was set using percentage as the unit of measure in the first place. What was the underlying reason(s) why this objective was chosen to reflect the performance of the procurement department?

I have to go back to understand the roles and responsibility of the procurement engineers. It says, ‘To ensure that the raw materials meet our product specifications and they do not cause any production downtime.’

With that I asked, ‘Why do you not set the objective as ‘number of incidences of non-conformance’? Certainly this would reflect the roles and responsibility of your department more accurately.’ There was complete silence.

I must have poked into their sore point - something that they had been successfully hiding all this while. They must have chosen to use ‘percentage of non-conformance materials supplied by the vendors’ to hide their poor performance.

Why do I say so?

For example, you have 100 cases of non-coming materials shipped to the production shop floor. Each time, a non-conforming material delivered to the factory would cause the production to stop immediately. Often it requires the production planners to re-schedule the production line to produce some other products in order to avoid precious production time lost due to the long downtime.

But changing the production schedule to produce another product means the entire production line has to be stopped for the setup or conversion to take place. This might take several hours to complete and possibly a day or two to ramp up the new set up to a smooth and efficient operation. Meaning, the factory would lose quite a lot of production output unnecessarily. All these occurrences that caused production downtime could be avoided if the procurement engineers had done a good job.

With a clever stroke of the pen, the procurement engineers white-washed their responsibilities away cleanly. Dividing the 100 cases of no-conforming materials over a much larger total number of deliveries of raw materials gave a resulting percentage that looked extremely small. In this case, the goal was set at 0.25%. The actual performance for the month of November 2005 was 0.19%. These two numbers were indeed small.

Quite naturally the reviewing party would not ask for an absolute zero value in the percentage. To achieve zero percent doesn’t make sense or possible to most people.

With that the procurement engineers had effectively put themselves in a very strong and defensible position NOT to achieve zero occurrences of non-conforming materials. This is another fine act of burying the head of the reviewing party deep into the sand.

In my point of view, certainly this would not be the case, if they were to measure their performance by the ‘number of occurrences of non-conformance’. At more than a hundred cases a year, the CEO would have jumped sky-high. The production line affected by a hundred times in a year is certainly unacceptable. Even once a week is unacceptable.

However, by cleverly setting the ‘right’ objective, the sore performance was effectively buried out of sight. Nobody makes noise about the non-conforming incoming materials. 

No wonder Hewlett-Packard’s vendors rated it as the best customer. Even when the vendors are producing non-conforming products, its procurement engineers were very comfortable with their rather poor performance. That means, practically little or no goods returned to the vendors for rework or to be scrapped at their expenses. Hewlett-Packard procurement engineers had effectively shielded them from the need to take corrective actions to remedy their poor quality.

But the reality in the production shop floor genuinely requires zero percentage of occurrences of non-conforming materials delivered to the production line! If the procurement engineers had stuck to the basic fundamental purpose of their department, they should set their objective as ‘number of incidences of non-conformance’ to truly reflect how best they had been helping the production lines to avoid unnecessary downtime.

Obviously, this objective tells the whole world that the ultimate goal is zero occurrences - this is what it should be.

What is the rationale behind moving away from this straight-forward and meaningful objective?

Information manipulation! Select the ‘right’ objective and truncate away as much information as possible. The procurement department had painted an excellent performance benchmark for the reviewing party. Bottom-line, there is no need to work hard to crack the whip on the vendors. This is buffer mentality to its hilt.”

“I could not imagine HP procurement engineers could be so skillful in deploying the art of using percentage to buffer against having to drive their vendors to deliver better quality products. These people really ‘work smart’. Hats off to them,” said John.


 

  评论这张
 
阅读(234)| 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