Improvement in a silo

In a discussion he recently started in the PEX Network discussion group on LinkedIn, Adi Gaskell asked whether process improvement worked in a silo. Most participants said no, but Steven Borris said yes, and I agree with him. Following is what I added:

I agree with Steven, and will even go further: your first pilot projects when you start Lean implementation have the best chance of success if they are contained within a department. The more departments, silos, or fiefdoms you involve, the more difficult you make it, and the less likely to succeed.

The scope does not have to include a complete process from raw materials to finished goods. It does not even have to be at the end or the beginning of the process. All his has to be is a process segment with a technical potential for improvement that is achievable with available skills, and enthusiastic local management.

There is a simple criterion to establish whether such a local project improves the plant as a whole: does it move its target operations in the direction of takt-driven production. If it does, and only if it does, the order-of-magnitude improvements you get locally translate to nibbling percentages globally. For example, the local WIP drops by 90% and that makes the global WIP drop by 4%.

Only once you have a few successful within-silo projects under your belt do you have the support in the organization and the skills base to take on cross-silo or silo-eliminating projects.

16 comments on “Improvement in a silo

  1. Comment in the PEX Network & IQPC – Lean Six Sigma & Process Excellence for… on LinkedIn:

    Improvement projects can be started at Organisational level, Functional level ( Finance, HR, Recruitment , Technical performance etc) & at a individual level

    The organisational level improvements themes cuts across the business units or lines of business . The Functional Themes would focus in improving the functions business outcomes & KPI’s. The Individual level improvement projects empower the team member to do improvement projects & inspire them to do more. Improvement projects scoped this way would address the needs at all levels. This has to be supported by the relevant Sponsorship & governance within the organisation depending on the theme & socpe chosen. Defining the Problem statement , the scope, the business or technical criticality would decide the length & breadth of the process & would mitigate risks of Siloed process improvement . This has worked reasonably well but an issue that has not gone away completely

    The measures are also important. The improvement project could Improve application performance or reduce cycle time but business would understand only Availability & Profitability. So how we identify the right measures for the rigtht audience would also make a difference

  2. Comment in the PEX Network & IQPC – Lean Six Sigma & Process Excellence for… on LinkedIn:

    What Ganesh and Michel refer to as possible in silo is underpinned and only made possible by the organization having a mature level of process management in place, i.e., process owners, design council, scorecards etc. If not the results scare me.
    Everyone agrees that any project that improves an aspect of accepted effectiveness and efficiency at the micro level will do little harm and will further the cause.
    However, I much prefer to target process improvement projects at the end to end customer experience level, ensuring all silos act as one.
    Great discussion

  3. Comment in the PEX Network & IQPC – Lean Six Sigma & Process Excellence for… on LinkedIn:


    Many companies have few KPI’s, no documented specs and the management is disconnected. This is why we have had to introduce some pilots in “willing” areas.

    I have been fortunate to work in world-class organizations and those at the other end.

    There are companies out there who are very good at what they do, but have few formal processes. Indeed, that is when they find the need to start looking for a new way to move forward.

    We have to start right at the beginning and baseline the project.

    Working in a specific area can be seen as an advantage to the team. They get a first hand opportunity to experience the people issues involved in an implementation.

    The task is not always simple, but we need to be flexible and adapt to the needs of the organization.


  4. Comment in the PEX Network & IQPC – Lean Six Sigma & Process Excellence for… on LinkedIn:

    Valid points. Perhaps I am biased due to the fact that I have seen too many projects launched and when you stand back they do not add to more than the sum of their parts. In instances silos have with all good intent harmed the customer relationship and cost the organisation time. effort and money to correct.
    But you need to start somewhere.

  5. Comment in the PEX Network & IQPC – Lean Six Sigma & Process Excellence for… on LinkedIn:


    I understand poor projects: I have seen the results of people carrying out SMED on equipment that feeds the bottleneck.

    To be fair, it was not the best lean practitioner.

    In what ways did your examples impact the customer?

    Could the problems have been avoided? Any advice?


  6. Comment in the PEX Network & IQPC – Lean Six Sigma & Process Excellence for… on LinkedIn:

    Need to be careful here, but for example – a bottleneck was causing delays in one operation, so after their silo review they decided to redirect volumes away from the bottleneck, made sense at first glance. However, they did not realise that downstream this would manifest itself into a major incident with customers (late deliveries).
    Parties that do not understand the end to end cause and effect if left to their own devices must be seen as a risk.
    Process improvement project selection needs to go through a process design authority first and only then launched (at least in my business – supply chain logistics)

  7. Comment in the PEX Network & IQPC – Lean Six Sigma & Process Excellence for… on LinkedIn:


    The process I follow is to have someone in the company become the lean leader or “champion” (I hate that word – can anyone suggest a better one?). I mentor him/her and they hold the fort when I am not in the company.

    It also helps to have a small group that approves projects and has the authority to clear roadblocks and authorise funding. It should have at least one senior manager on the team should hold (semi-formal) progress meetings, which helps to improve sustainability.


  8. Comment in the PEX Network & IQPC – Lean Six Sigma & Process Excellence for… on LinkedIn:

    I am developing a version of CEMM, and use it to focus everyone on how silo processes interact with one another to deliver;
    Moment of Truth – a touch point with the customer that has a clear expectation on performance (effect)
    Break Points – that cause of the MOT to fail
    Business Rules – the why

    The start point in the area of improvement is to firstly understand the above and then and only then develop solution ideas. Very simple yet effective

  9. Comment in the PEX Network & IQPC – Lean Six Sigma & Process Excellence for… on LinkedIn:

    Great discussion, folks . . . Can process improvement be accomplished in a silo? I would say yes, if it is a small, quick-strike improvement that either has a positive or no impact on updownstream and downstream processes. That being the case, a single silo’s improvements are usually trivial and not taken seriously by the rest of the organization. This works well when all silos are engaged to a point where there are no silos – people working together.

    Also, when these silo improvements are incorrectly implemented, they may fix an internal problem in a silo but create additional problems for others in the cross-functional process.

    When it comes to serious improvement and culture change, too many people look for the quick and easy way, the magic recipe. Continuous and sustainable improvement is logical and looks easy, but it is not. The only way to make a large competitive impact is to step up and do it right – plan, deploy, execute, and sustain a “continuous” improvement process across the entire enterprise. No one should be exempt from improvement in this challenging economy.

  10. Many of the comments express the concern that a local project may actually make things worse for the plant as a whole. Obviously, no matter what it consists of, such a change does not deserve to be called “improvement.”

    The concept of takt-driven production, however, gives you criteria to make sure that what you do in one silo is not hurtful elsewhere. Assuming you have to deliver your finished goods at a given takt time to meet demand within your work time. Then takt-driven production is the ideal situation in which each unit of product transfers instantly to the next operation at every beat. It is as if you had a metronome set to the takt time, and every time it ticks, each product unit pops forward.

    Takt-driven production is an ideal that would be realized globally by realizing it locally everywhere. Of course you never realize it, but you approach it by eliminating waste. When you examine a proposal for a local project in a plant, it is not too difficult to tell whether it is moving this section of the plant in this direction or not. If it does, you can rest assured that this project will not make conditions worse elsewhere in the plant. Conversely, if it moves in the wrong direction, you can be sure it will and you should not support it.

  11. I think you need to start this conversation with a definition of the type of silo you are referring to. If you are referring to a process silo like Paintshop or Metal Stamping, then I agree with you. Of course you can make improvement in a small area as long as you have a clear ideal state that the improvements can be tied to. There is always a threat of sub-optimization but everyone is a silo to some degree or another. If you were to say that you can’t improve in a silo, you would essentially be saying that you can’t improve unless you own every step of every process from mining from the ground to selling to the final customer.

    When I think about silo’s however, I am usually thinking more about the functional and organizational silo’s. The big three in most traditionally managed companies are Operations, Engineering, and Maintenance. Those are the silo’s that kill improvement. If I think of these silo’s, I don’t really agree. It will at lease be very difficult to make SUSTAINABLE improvement without representation from key stakeholders across functional silo’s.

    It sounded to me like half the people in these comments were talking about one type of silo and the other half were talking about another.

    • Things you can do in a paintshop or a stamping plant include reducing setup times to cut batch sizes and reduce your “every part every” interval. This you can do internally, without involving anybody else, except possibly some support from maintenance.

      Then you can organize your output buffer as a supermarket to allow downstream processes to pick from it in heijunka sequence, but that requires the involving the downstream process. Likewise, in stamping, you need to work with the downstream process to use the signal/material kanban combination to trigger the retrieval and preparation of the next dies before kicking off each setup.

  12. Agree with Steven and one thing more to add. Attraction works well in any kind of change – especially the dramatic changes that lean can bring to an organization. Promotion does not work well – people feel like they are having something “sold” to them or forced on them.

    The implementation within one department or area gives the opportunity for the others in the surrounding areas to observe. They watch and learn from observing the transition in the other area. This is the next best thing to experiencing it for themselves.

    But this last point brings to mind a critical part of the transformation in the one area. It must be heavily documented with good visual communication. Before and After data, graphs, photos. Displayed in the area – not in some Powerpoint show that is shown once or twice to management. A3’s are the best for this.

    ROBs (Report Out Boards) or Visual Walls in the area also make this much easier. When I do a transformation – I almost always make sure this is one of the first things to go up in the area. Usually a 4′ x 8′ white board. with To display the status and ongoing issues and results. And to start the group into reporting out daily to management on a Gemba walk to the area.

  13. Would you mind if I quote a few of your articles as long as I provide credit and sources back to your website: http:
    // I’m going to aslo make sure to give you the appropriate anchortext link using your webpage title: Improvement in a silo | Michel Baudin’s
    Blog. Please be sure to let me know if this is acceptable with you.
    Thank you

Leave a Reply

Your email address will not be published. Required fields are marked *