About Michel Baudin

Consultant, trainer, author on Lean Manufacturing

Posts by Michel Baudin:

Mining operations

Oct 26 2011

Data Mining in Manufacturing versus the Web

Data mining, in general, is the retrieval of information from data collected for a different purpose, such as using sales transaction histories to infer what products tend to be bought together. By contrast, design of experiments  involves the collection of observations for the purpose of confirming or refuting hypotheses.

This perspective on data mining is consistent with the literature in expressing purpose, but most authors go further. They include in their definitions that data mining is done with computers, using large databases and specific analytical tools, which I think is too restrictive. The tools they list are the ones they have found useful in analyzing the behavior of millions of users of search engines or commerce websites, and they are not obviously applicable in other areas, such as manufacturing.

During World War II, British analysts used the serial numbers of captured or destroyed German tanks to estimate the numbers produced. Because serial numbers were not attached for this purpose, it was data mining. It used clever statistical models but, obviously, no computers.

Today, PhD-level data miners at Google, eBay, or Amazon sift through the  page views and click-throughs of millions of users for clues to patterns they can use. The data, automatically collected, is accurate and collected by the terabytes every day. This “big data” requires parallel processing on clusters of computers and lends itself to the most advanced analytical tools ever developed.

Compared to this fire hose of data, what manufacturing produces is a trickle. In a factory, the master data/technical specs, plans and schedules, status of operations and work in process, and the history of production over, say, 12 months, usually adds up to a few gigabytes. It doesn’t  fit on one spreadsheet, but it often does on a memory stick. On the other hand, much of it is still manually generated and therefore contains errors, and it is often structured in ways that make it difficult to work with.

Even if manufacturing companies could hire the data miners away from their current jobs, their experience with e-commerce or web search would not have prepared them well for the different challenges of manufacturing data mining.

There is an opportunity for data mining to contribute to competitiveness in manufacturing, but the approach must start from the needs. It must not be an e-commerce cure in search of manufacturing diseases.

By • Technology • 0 • Tags: , ,

Steam locomotive and typewriter

Oct 25 2011

The steam locomotive and the typewriter

The first draft of my book  Working with Machines contained a chapter that was a post-mortem on two obsolete machines, which was cut on the grounds that, unlike all other chapters, it was not actionable for the reader.

Its abstract is as follows:

The steam locomotive and the typewriter are icons of the industrial age, and their parallel histories show different aspects of the human experience of working with machines. The steam locomotive is fondly remembered; the typewriter, all but forgotten except for the QWERTY keyboard. The steam engine participated in the development of every industrial economy, but the typewriter played no major role in Japan. The typewriter did not demonstrably improve the productivity or quality of office output, but was adopted only because of its image of modernity.

Locomotive driver was a prestigious position for a manual laborer, but typist never was. Compared to electrics and diesels, the steam locomotive had a cab that was exposed to the elements and to the heat of the firebox and therefore uncomfortable, difficult to operate, and dangerous. Yet engineers and firemen preferred it to the tedium and loneliness of modern locomotives. Automatic machines that require human attention only when they malfunction are also in airplanes and in manufacturing plants, challenging the job designer to keep the operator alert and used efficiently.

As the typewriter prints one keystroke at a time, typists were always busy with a single machine and determined both its productivity and output quality. Typists worked in comfortable places, but under pressure, and faced the long-term hazards of sedentary work. The typewriter’s main legacy is that a society can make a long-term investment in machines whose tangible benefits do not obviously exceed their costs.

Click here for a pdf file of the entire chapter.

By • History • 1 • Tags: , , ,

Manual data collection at end of shift

Oct 24 2011

A management perspective on data quality

Prof. Mei-chen Lo, of National University and Kainan University in Taiwan, worked with Operations Managers in two semiconductor companies to establish a list of 16 dimensions of data quality. Most are not parameters that can be measured and should be considered instead as questions to be asked about a company’s data. I learned it from her at an IE conference in Kitakyushu in 2009 and found it useful by itself as a checklist for a thorough assessment of a current state. Her research is about methods for ranking the importance of these criteria.

They are grouped into four main categories:

  1. Intrinsic. Agreement of the data with reality.
  2. Context.  Usability of the information in the data to support decisions or solve problems.
  3. Representation. The way the data is structured, or not.
  4. Accessibility. The ability to retrieve, analyze and protect the data.

Each category breaks further down as follows:

  1. Intrinsic quality

    • Accuracy. Accuracy is the most obvious issue and is measurable. If the inventory data says that slot 2-3-2 contains two bins of screws, then can we be confident that, if we walk to aisle 2, column 3, level 2 in the warehouse, we will actually find two bins of screws?
    • Fact or judgment. That slot 2-3-2 contains two bins of screws is a statement of fact. Its accuracy is in principle independent of the observer. On the other hand, “Operator X does not get along with teammates” is a subjective judgment and cannot carry the same weight as a statement of fact.
    • Source credibility. Is the source of the data credible? Credibility problems may arise due to the following:
      • Lack of training. For example, measurements that are supposed to be taken on “random samples” of parts are not, because no one in the organization knows how to draw a random sample.
      • Mistake-prone collection methods. For example, manually collected measurements are affected by typing errors.
      • Conflicts of interest. Employees collecting data stand to be rewarded or punished depending on the values of the data. For example, forecasters are often rewarded for optimistic forecasts.
    • Believability of the content. Data can unbelievable because it is valid news of extraordinary results, or because it is inaccurate. In either case, it warrants special attention.

  2. Context.

    • Relevance. Companies often collect data because they can, rather than because it is relevant. It is the corporate equivalent of looking for keys at night under the street light rather than next to the car. The semiconductor industry, that established this list of criteria, routinely takes measurements after each step of the wafer process and plots them in control charts. This data is relatively easy to collect but of little relevance to the control and improvement of the wafer process as a whole. The engineers cannot capture most of the relevant data until the circuits undergo tests at the end of the process.
    • Value-added. Some of the data produced in a plant have direct economic value. Aerospace or defense goods, for example, include documentation of their production process, as part of the product. More generally, the data from commercial transactions, such as orders, invoices, shipping notices, or receipts, is at the heart of the company’s business activity. By contrast, the organization generates data to satisfy internal needs, including, for example, the number of employees trained in transaction processing on the ERP system.
    • Timeliness. Is the data available early enough to be actionable? A field failure report on a product that is due to problems with a manufacturing process as it was 6 months ago is not timely if this process has been the object to two engineering changes since then.
    • Completeness. Measurements must come with units of measure and all the data describing who collected them, where, when, and how.
    • Sufficiency. Does the data cover all the parameters needed to support a decision or solve a problem?

  3. Representation

    • Interpretability. What inferences can you draw directly from the data? If the demand for an item has been rising 5%/month for the past 18 months, it is no stretch to infer that this trend will continue next month. On the other hand, if a chart tells you that a machine has an Overall Equipment Effectiveness (OEE) of 35%, what can you deduce from it? The OEE is the product of three ratios: availability, yield, and actual over nominal speed. The 35% figure may tell you that there is a problem, but not where it is.
    • Ease of understanding. Management accounting exists for the purpose of supporting decision making by operations managers. Yet the reports provided to managers are often in a language they don’t understand. This does not have to be, and financial officers like Orrie Fiume have modified the vocabulary used in these reports to make them easier for actual managers to understand. Engineers using cryptics instead of plain language make technical data more difficult to understand.
    • Conciseness. A table with 100 columns and 20,000 rows with 90% of its cells empty is a verbose representation of a sparse matrix. A concise representation would be a list of the rows and columns IDs with values.
    • Consistency. Consistency problems often arise as a result of mergers and acquisitions, when companies mash together their different data models.

  4. Accessibility

    • Convenience of access. Data that an end-user can retrieve directly through a graphic interface is conveniently accessible; data in paper folders on library shelves are not. Neither are databases in which each new query requires the development of a custom report by a specially trained programmer.
    • Usability. High-usability data, for example, take the form of lists of property names and values that you can easily tabulate into spreadsheets or database tables. From that point on, you can select, filter and summarize it in a variety of informative ways. Low-usability data often comes in the form of a string of characters that you first need to parse, with character 1 to 5 being one field, 6 to 12 another, etc. Then you need to retrieve the meaning of each of these substrings from a correspondence table, to find that ’00at3′ means “lime green.”
    • Security. Manufacturing data contain some of the company’s intellectual property, which you need to protect not only from theft but from inadvertent alterations by unqualified employees. But you must also provide security effectively so that security procedures do not slow down qualified, authorized employees accessing data.

Prof. Mei-Chen Lo’s research on this topic was published in The assessment of the information quality with the aid of multiple criteria analysis (European Journal of Operational Research, Volume 195, Issue 3, 16 June 2009, Pages 850-856)

By • Management • 4 • Tags: , , , , , ,

Dashboard assembly stations

Oct 23 2011

Why production matters

On LinkedIn Lean Business Process group, Ralph Bartelmann asked the following:

Is it really the matter to squeeze the last cent out of production ? In many environments production costs represent a minor part of the over all product cost. Following Pareto reasoning it seems more reasonable to work on other parts of the value stream like supplier developpement, product design etc. What is your opinion and experience ? What are the real challenges ?

Following is my answer:

It’s not about what production costs  but about what it does for the business. Improving production is about making is faster, better, safer, less tedious,… and cheaper.  It needs to be faster to make you more responsive, better so that production does not introduce defects that harm your reputation, safer and less tedious so that you can retain your work force and grow its skills.

If you improve on all these fronts, guess what? Your costs go down, and not only in production but in other parts of the value stream too, because they are not independent of production. For example, there is no point in trying to develop just-in-sequence suppliers unless you practice leveled-sequencing (a.k.a. Heijunka)  in your assembly line.

A manufacturing company ignoring production is an army ignoring combat on the grounds that more money is spent moving soldiers and keeping them supplied.

By • Management • 0 • Tags: , ,

Oct 22 2011

This is a poll for managers and engineers

This is a poll for managers and engineers working in factories: how to you use manufacturing data? Please answer on http://linkd.in/ogOsOv, and comment if none of the choices fit you.

By • Polls • 3 • Tags: , , , ,

To get to Dublin, you don't start from here

Oct 22 2011

Kaizen without prerequisites

Question from Pedro Burgos: You have an assembly line where ‘Standardized Work’ is not utilized. The operators have ‘work instructions’ and follow them. They do not miss steps. But there are differences between operators in how the work is performed in terms of sequence. All of them perform under takt time, some more than others. Due to ‘wastes’ of motion. You are asked to do a Kaizen to reduce inventory quantities and space. Inventory quantities on the production floor varies, some parts have 3 days of inventory, some have 2 days, some have 1 day (16 hours). The goal is a ‘kanban’ of 2 hours of inventory in the assembly line. Do you think is possible to do this Kaizen? (Discussion in the LinkedIn Global Lean & Six Sigma Network)

Your question reminds me of the old story about asking a farmer in the countryside of Ireland how to get to Dublin, and getting the answer that “you don’t start from here.” In fact, regardless of its current state, a production line can always be improved.

Looking further at the way you describe the challenge, however, several things disturb me. First, while Kaizen, meaning improvement, is always possible, “a Kaizen” seems to designate a Kaizen event, a project template which is not always appropriate. Second, your description suggests opportunities for productivity and quality improvement, but neither is stated as an objective.

A top-down mandate to use a particular method to pursue two goals like inventory and space reduction does not strike me as a recipe for success. Instead, someone needs to do the following:
1. Take a close look at this assembly line.
2. Identify improvement opportunities in all dimensions of its performance.
3. Select one or two to pursue in priority based on improvement potential, technical feasibility, and human feasibility.
4. Determine the project management approach based on content, rather than the other way around.

This approach is much more likely to achieve the desired inventory and space reductions, among other improvements, than just running a Kaizen event to pursue these objectives directly and exclusively.

By • Management • 1 • Tags: , , ,

«< 161 162 163 164 165 >