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.
Oct 27 2011
Waiting for each other
We have all seen the absurd situation in the featured picture above of a line of customers waiting for taxis while a line of taxis next to them is waiting for customers, with a barrier separating them. This particular instance is from The Hopeful Traveler blog. The cabs are from London, but the same scene could have been shot in many other major world cities.
I am sure we have all encountered similar situations in other circumstances, which may or may not be easy to resolve. One particular case where it should be easy is the restaurant buffet. Figure 1 shows a typical scene in buffet restaurants, with a line of people waiting to get food all on the one side of the table, while food is waiting and accessible on the opposite side.
Figure 1. A typical buffet
I think the fundamental mistake is the assumption that a buffet is like an assembly line, providing sequential access to dishes. This means that you cannot get to the Alo Gobi until the person in front of you is done with the Tandoori. The ideal buffet would instead provide random access, meaning that each customer would have immediate access to all dishes at all times. While it may not be feasible, you can get much closer to it than with the linear buffet. The following picture shows an alternative organization of a buffet in circular islands that is non-sequential.
Figure 2. A buffet island at the Holiday Inn in Visalia, CA
The limitation of this concept is that replenishment by waiters can interfere with customers. To avoid this, you would want dishes to be replenished from inside the circle while customers help themselves on the outside, as in the following sketch:
Figure 3. A buffet island with replenishment from inside
One problem with the circular buffet island, however, is its lack of modularity. You can add or remove whole islands but you cannot expand or shrink an island, which you can if you use straight tables arranged in a U-shape, as in Figure 4.
This buffet island may superficially look like a manufacturing cell, but it is radically different. Its purpose is random access to food as opposed to sequential processing of work pieces, and the materials do not flow around the cell but from the inside out.
Such are the thoughts going through my mind while munching on the Naan at Darbar.
Share this:
Like this:
By Michel Baudin • Technology • 3 • Tags: Buffet, industrial engineering, Lean manufacturing, Restaurant