4 reasons to consider SMED

SMED is a structured approach to reducing changeover durations. Here are 4 good reasons to consider deploying SMED.

>Lisez-moi en français

1. More production capacity is required

A critical resource of the production process is not able to deliver the expected quantity, due to lack of capacity. Such a resource is usually called a bottleneck, and chances are that a quick analysis will show that changeover durations are one of the major factors of capacity wastage.

Rather than reducing the number of changes and extending the series, it’s better to work on reducing the duration of changeovers from one series to the next in order to recover wasted capacity.

In many cases the gains through the application of the SMED method are sufficient to recover the lacking capacity. If it’s not the case, it is then necessary to check the distribution of the causes of capacity wastage (via a Pareto diagram, for example). Machine downtime due to breakdowns or quality issues discarding a part of the production may be additional themes to deal with.

2. More flexibility is required

When production capacity is sufficient but there is a need to better stick to the demand, this can be achieved by changing the productions more frequently.

Indeed, for a customer waiting for a given product, the waiting time is function of the length of the queue of orders to process before launching the production of the expected product, the lead time of the production process itself, and finally the shipment process.

The reduction of the batch sizes and the multiplication of launches help reducing the global lead time. For this solution to be viable, however, it is necessary that the multiplication of changeovers does not cause the production capacity to fall below the required level, otherwise we find ourselves in the previous case.

To prevent this risk, the duration of the changes can be reduced by applying the SMED method and converting the time gained into additional changeovers.

3. More machine availability is required

In this third case there is no shortage of production capacity nor lack of flexibility but what is lacking is time during which a machine is available for periodic maintenance operations, or for processing exceptional additional orders.

In this case also the reduction of the duration of changeovers may be a solution to consider. The recovered machine availability may then be used as needed.

4. Freeing time of experts

This case is rare, but may arise, especially in environments that are strongly constrained by standards or regulations and in which changeovers require the assistance of personnel with special qualification and/or authorization. In these cases it is no longer the capacity or the availability of the production means which is limited but those of these “experts”.

They may be required simultaneously for different changeovers and their limited availability may adversely affect the overall performance of production, with some machines, equipment or lines waiting.

Take care of adjustments and tests too

The term “changeover” is sometimes interpreted as the change of dies, or the change of settings only. A changeover should include the clearance of everything related to the completed series and the resuming of production with the new series, new reference or new batch, at nominal speed. This means that all required adjustments and tests have been performed and their durations have been counted as part of the changeover.

Excluding these durations in order to pretend to change over quickly may leave them out of the scope of improvements, which is a loss of improvement opportunity and only a partial application of SMED.
View Christian HOHMANN's profile on LinkedIn

Advertisements

Smart Factories: Converting workers into problem solvers?

About the author, Chris HOHMANN

About the author, Chris HOHMANN

Smart factories, industry 4.0, IIoT and everything else covered by the digital transformation umbrella raises fear about the future of human jobs, or to be more specific, the income source of employees.

To my surprise some promoters of these factories of the future have a ridiculously simple solution: convert the actual workers into process supervisors and problem solvers, without questioning the paradox of the need for supervisors and problem solvers when processes will be – allegedly – way more reliable, constant, repeatable and even self improving.

>Lisez-moi en français

Even so the assumption of required supervisors and problem solvers is valid, I do believe that the number of freed workers through automation, robotics, etc. will exceed the number of required supervisors and problem solvers.

If it would not be the case, it is the benefits of the promoted solutions that are questionable. Especially if many more problems solvers are required, it should raise concerns.

Besides, as I am not impressed by most of the actual, traditional manufacturing problem solvers’ capabilities and achievements, I cannot believe they will do much better with technologies requiring even keener knowledge.

Am I wrong?

Am I the only one perplexed by this “solution” for converting the redundant people?

Your thoughts are welcome.

Why you cannot use tentative language in a logic tree

I once happen to see a Current Reality Tree cluttered with “coulds” and “shoulds”. Conditional or tentative language cannot be used with logic trees and here is why.

Cause-and-effect (sufficiency logic)

The Logical Thinking Process logic trees use either sufficiency or necessity logic. Sufficiency or cause-and-effect relationship states that a cause, if it exists, is sufficient by itself for the effect to happen. Using conditionals like should or could violates the sufficiency principle as it suggests that the cause is not always producing the effect.

The Current Reality Tree (CRT), Future Reality Tree (FRT) and Transition Tree (TT) are built on sufficiency logic and therefore cannot hold any entity with shoulds or coulds.

If a should or could is found in such a tree, the scrutinizer must raise a “cause insufficiency reservation“. The statement must then be corrected, for example by adding one or more additional cause(s) combining to the first one with a logical AND connector. If this combination is valid, the sufficiency relationship is restored and should or could is removed as the effect is now guaranteed to happen.

If no additional causes can combine to the first one, the cause-and-effect relationship is probably only assumed or false. Anyway no should or could can be left in a logically sound tree.

Using present tense

The entities – the building blocks of the logic trees holding the statements – must be expressed in present tense.

Using present tense is natural in a Current Reality Tree (CRT) as it is the description of the actual situation, the cause-and-effects relationships that exist right now.

The use of present tense in Future Reality Trees (FRT) is highly recommended even so these future situations and the Desirable Effects do not yet exist. Present tense helps to project oneself and the audience into the future and visualize the situation as it were already improved (Scheinkopf, “Thinking for a change, putting the TOC Thinking Processes to use”, p119). Dettmer also recommends to use positive wording (Dettmer, The Logical Thinking Process, p244).

This applies to entities in a CRT, a FRT and in a Prerequisite Tree (PRT) which are verbalized in full sentences.

What about necessity-based logic?

Can necessity logic based tree use conditional/ tentative language?

The Goal Tree (GT), the Evaporating Cloud (EC) and Prerequisite Tree (PRT) are built on necessity logic. They describe the chains of enabling conditions that are required to achieve a goal or an objective. Without the enabling conditions, the objective cannot be attained. Conversely, with the enabling, necessary conditions fulfilled, the objective will not automatically be achieved; additional action is required.

As the Desired Effect is not guaranteed to happen even so all necessary conditions are fulfilled, the use of conditional / tentative language seems legit. Practitioners would not use it though.

First because we need to demonstrate positivity about a desirable change and help the audience to mentally visualize the future where things happen and produce the desired outcome.

Second because we need to give confidence and demonstrate our own trust in the proposed solution. No audience would be thrilled hearing that this solution “may”, “should” or “could” produce the desired result. No decision maker would give his/her go for a change program or a solution implementation which is not certain to produce the expected result.

The use of conditional / tentative language would only raise concern about the feasibility of the proposed solution and appear as a lack of confidence of its promoters.

Wrapping up

Tentative language is recommended in academic writing, not at all with logic trees.

Using tentative language is recommended in academic writing and scientific research in order to leave room for alternatives, later corrections, etc. unless there is solid evidence backing up a statement. Therefore the use of verbs like “appear, suggest, indicate,…”, modals “may, might, can, could, will, would” and adverbs like “possibly, probably, likely…” are recommended.

But when building or presenting logic trees, absolute certainty is required in order to demonstrate robustness of the analysis and the confidence in the conclusions. If a logic tree is built on the canonical logic rules (we’ll consider the use of present tense as a canonical logic rule), has been scrutinized and cleared of all reservations, it is robust and tentative language is no option.


The author, Chris HOHMANN

The author, Chris HOHMANN

View Christian HOHMANN's profile on LinkedIn

The Logical Thinking Process – An Executive Summary (Book)

Bill Dettmer, my friend and mentor often cited on this blog, wrote his 9th book “The Logical Thinking Process – An Executive Summary” that is now in the final stages of the publishing process.

This book will be much smaller in size and number of pages than the famous “The Logical Thinking Process: A Systems Approach to Complex Problem Solving”. The latter is a 413 pages, 17,1 x 3,2 x 24,1 cm (6.8 x 1.2 x 9.5 inches) hardcover book. Bill Dettmer’s students use to call it “The Bible”. It is a complete step-by-step guide, easy to read and understand, but not everybody can invest the time required to read it just to get a primer on the Logical Thinking Process.

That’s where the “Executive Summary” comes in handy. Bill himself stated: “Over the years, I’ve found myself having to explain what the Logical Thinking Process is in 30 seconds or so to people who have never heard of it – or know nothing about it if they have. I came to the conclusion that while the LTP is difficult, if not impossible, to encapsulate in an “elevator speech,” it might be somewhat easier to do in a pocket-sized book.

I was fortunate to be selected by Bill as a kind of sounding board and proofreader, even so Bill did so more out of kindness than necessity, and had a privileged first reading. The book will serve its purpose, I think it can be read during a short business trip on a plane or a train. The final copy should be less than 100 pages.

As soon as I’ll get a copy, I will complete this post with an extended review of the book.

You may subscribe to my blog to keep being informed of new posts or follow me on twitter.

View Christian HOHMANN's profile on LinkedIn