APS ideas Blog

Which comes first: Scheduling or Shop Floor Data Collection?

This article was co-authored with John Leibert of Factivity, www.factivity.com.

Manufacturers are faced with many choices when it comes to improving the flow of materials and using their constrained resources more effectively.  Planning and scheduling, and collecting production data from the factory floor are a part of running just about any manufacturing business.  These activities are especially important for manufacturing companies that want to use information systems to shrink inventory levels, improve on-time customer delivery, and increase productivity.  Using these systems as one, is a solid approach to becoming more lean and world-class in the production process.

Scheduling using tools such as spreadsheets, ERP reports, production meetings, and phone calls gets results - up to a point.  Likewise, collecting production data on handwritten forms, and entering it into the ERP system or an accounting database might be perfectly acceptable too, at least for a while.

When both the ability to schedule and the ability to collect good factory floor data appear to be in jeopardy, which should be improved first?  This question might have more than one answer, so let’s explore some ideas and options.  A realistic, rules-based leveling loading scheduling system (APS) could be a good idea.  An easy-to-use, real-time shop floor data collection system (SFDC) also provides benefits.  In the end, it’s going to come down to which system will provide the most improvement and therefore  be the project that drives the most value, the soonest.  If only it were that easy to determine! 

Conventional wisdom says that because scheduling is so “data intensive”, that a factory floor manufacturing system (MES) or a shop floor data collection system should be in place first so that data is accurate and timely for updating the schedule.  But what if you choose to improve scheduling first, while SFDC is allowed to continue as-is for a little longer?

-          The initial scheduling system project should provide

o   a quicker and

o   more efficient and

o   more realistic

day-to-day schedule for all factory floor resources than the current approach.   It is true that this new system is restricted to using the same product (routing time standards) and timely production data collection at the start.  Remember…your previous approach was stuck with the same data quality, so shouldn’t a better system provide some forward progress?

One theory is that if you select a scheduling system that is better for your business overall, you are automatically ahead, even if your part/production data remains in the same state it is in today. 

Let’s look at the opposite approach.  What is the impact of focusing a new project to improve your part/production data by installing a new shop floor data collection system?  How does improved SFDC work if scheduling is not improved?

-          Timeliness of information should increase

-          Daily production decision-making might be better

-          Measurements of “what happened” and “who did it” are more trustworthy

If your company is struggling with knowing “what happened”, SFDC should be a priority. Depending on your implementation of a factory floor system,  your company should have a factory floor system that provides management with measurements of labor and equipment effectiveness and up-to-the-minute job tracking.  A more advanced system may also help improve routing time standards, make it easier to share processes and procedures documents, and deliver useful alerts to problems.  Again, in certain areas, your company will come out ahead.

Beyond the benefits, the amount of time, effort and cost to implement the project will be a major consideration.  Depending on the focus, your resource involvement will vary.

For example, with a new scheduling system, people from Production Scheduling will be involved.  And it is likely that the IT department will be involved, depending on the type of interfacing to other internal business systems (such as ERP).  Hardware may require upgrading.  For scheduling systems, a new high end PC/Server with a wide screen and high resolution is normally a requirement. 

Installing an SFDC or MES system requires more equipment and therefore more cost for deployment on the factory floor.  In addition to the floor labor force, supervisors and some office staff (production) might need access to the floor system.  You should expect IT and production personnel to be involved directly with the training and implementation.  Accounting, manufacturing, engineering, quality and even upper level plant management may be involved.  Implementing an SFDC requires good project management and involvement of more departments in the project.

With an APS, the scheduler is committed and production is involved; with an SFDC/MES the entire floor is committed and the scheduler is involved.

Because any change involves cost, time, effort, and risk to the business, the best way to be sure of your choice is to lay out all of the options clearly.  This often means working closely with the companies that might have a potential solution for you.  They will be most familiar with the pros and cons of their own approaches, if not many of the alternatives as well.  Where would you focus your improvement efforts today?

Successful APS Implementations, Part 2: Working through a Software Project

So, you’ve read through Part 1 and are feeling prepared to change the way you are managing your production schedules and plans.  That’s great!  Now, building on those preparations and seeing the project through to a successful completion is the next step.

This is where the expertise and guidance of your implementation consultant will be really helpful.  Consultants that listen well and can see how your specific needs are similar (or not) to their previous customers are invaluable.  In addition to providing valuable advice on the details, your consultant should be able to help manage the project from beginning to end.  With your team, this involves defining the scope, listing the milestones to achieve and the deliverables needed along the way.  A big part of defining the scope is separating the “must-haves” from the “nice to haves”.  In other words, prioritizing your time and resources on the most vital aspects of the requirements first.

Within this structure, knowledge and information is shared both ways between consultant and the project team.  Within the team, the “Subject Matter Experts” (SMEs) are absolutely critical.  With scheduling software implementations, the SMEs are usually the master schedulers – and that means they are usually already scrambling to keep up with their normal workload.  The master schedulers need to be present for the early discovery sessions, and during the factory modeling prototypes that drive the system integrations.  When that’s all done, let’s not forget about training and education along the way.  Whew!  Let’s back up a little…

Starting with the discovery and interview process, your consultant will be able to set a course for how your APS system will be implemented.  Most of the time, APS systems are heavily dependent on the availability of data (and those systems would work best with accurate data.)  ERP, inventory management, and order management systems come in all varieties, and typically have a majority of the data needed to go into an APS.  Connecting to that data is a technical job, driven by business requirements.  Factory modeling and prototyping are easy ways to validate ideas before committing technical people to developing an interface between your systems and an APS.

Your master schedulers need to be involved throughout the project, validating ideas as you go.  They will get some exposure to the new APS as it is designed, but it is critical to include separate time for training and education.  Many times, using an APS is very different from relying on a spreadsheet to number-crunch, and relying on a master scheduler to make the decisions.  The transition to an APS is easier when discovery is done first; then prototyping; then integration; and then formal training tailored to the scheduler’s responsibilities.

Having a status meeting each week helps keep your team and your consultant accountable for tasks, and the time and money needed to allocate your project resources. It is important to be able to make adjustments to the project along the way, as needed.

Conference room pilots are a must, and with good procedural documentation along the way the pilot should lead to a smooth “go-live” event.  Then it’s time for a well-deserved celebration before the next scheduling crisis hits – but now you should be ready with a system designed to optimize your manufacturing schedules, working hand-in-hand with your trained experts to handle any situation quickly and effectively.

Successful APS Implementations, Part 1: How to Prepare for a Software Project

I’ve worked with some great consultants over the years, and I thought that I would share some of the main tips and ideas our senior implementation consultants have found to be helpful when implementing a manufacturing scheduling software system.  This article will address the preparations you and your planning team can take to increase your chances of a successful project.  Part 2 will cover topics related to actually making it happen.  Of course, this assumes that you have gone through a rigorous evaluation process and have found the right system and company to work with for your needs.

I know it is a bit clichéd, but having an Executive Sponsor for any software project is still a key success factor.  Ideally, this Executive and the entire company have a vested interest in seeing the project through to completion.  Accountability for the whole project team needs to be in place, for both manufacturer and consultant, to keep the implementation on track.  The best Executive Sponsors are active “cheerleaders” for the team, keeping everyone positively motivated when the going gets tough and providing time when needed.  Everyone has a million things to do in their workday, and a strong sponsor brings much needed focus to any project.  It might go without saying, but true accountability means rewards for a job well done, and corrective actions when needed.

The Executive Sponsor does not need to be involved in the day-to-day details of the implementation.  That’s where the Project Manager comes in.  Throughout the implementation, a Project Manager may need to spend between 20-40% of her time during the week to ensure tasks are getting done.  Yes, that’s at least 8 hours a week and more at various times to make sure things happen.  Many times, the Project Manager is also a subject matter expert or an end-user, but not always.

Depending on their roles, other team members might be needed to provide their expertise along the way.  In general, setting aside at least 1 hour per day for every project team member is the best way to ensure consistency and maintain the momentum during the change to a new scheduling system.  Having this discipline and management in place when your APS consultant is not present is one of the key success factors, leading to projects completed on-time and on-budget.

In summary, making a change as significant as implementing a new way of scheduling your resources and materials requires a commitment.  That means people need to have time to spend, to share their expertise, and do the tasks needed.  The best way to ensure that is to have top-down support and a good project plan in place before embarking on a new project. 

If the time and resource commitments seem daunting, think about other challenging learning experiences you have been through.  How long did it take to create the scheduling program you have now?  How long did it take you before you felt like an expert at your career?  Dedicating an hour or so per day for several weeks to deliver lasting results can be a worthwhile investment if your company has the will to succeed.

Manufacturing production schedule compliance

“I would be interested in discussion about the process of following the recommendations provided by APS. Do planners load according to the plan, and if not, what are the reservations. And should they?”

One of the members of the LinkedIn group, “Advanced Planning & Scheduling (APS)” asked me this question, and I’m hoping to see that start as a discussion there and on other related groups.  This question is interesting because it asks if the users of a planning and scheduling system should provide the generated results to the plant.  I’m used to hearing whether or not the floor should follow the schedule, and this question happens a bit earlier in the process.

Human planners are able to analyze complex information and make good decisions based on their expertise and experience.  It’s much harder for a computer to do that, though IBM’s Watson supercomputer could be bringing the future that much closer.  Right now, I don’t see any full replacement for a human knowledge-expert, especially regarding planning and scheduling.  However, APS software systems have come a long way in terms of collecting data, sequencing and allocating tasks, and presenting information to help those expert users make the best decisions.

In short, it would be rare for an APS system to be completely “hands-off.”  However, these are the main elements that automate the plan creation most effectively:

1.       Thinking about the business problems you are trying to solve.  This has an effect on the design and use of the system.  Is the system for “long-range planning”, “short-term scheduling”, or both?

2.       Design of capacity and material constraints to allow the APS to make basic choices.  Examples: What should the APS do if the preferred resources are not available?  How should a potential material shortage be handled?  How should a new order be added to the plan?

3.       Design of the production flow, typically done using “routings” and bills of material/recipes.

4.       Deciding which parameters should affect the automated process that creates the schedule.  For example, what is most important to the company?  Meeting due-dates?  Reducing inventory?  High resource utilization?  Each of these objectives can conflict with one another.

If the constraints and “rules” are reasonably close for typical situations, then an average day could be mostly “hands-off” for the planning team.  Usually, something will happen that affects the schedule that cannot be planned for.  That’s where using “what-ifs” and helpful analytical tools make a difference in the company’s ability to react to unforeseen problems.  The speed and responsiveness of the APS will play into this too; if it takes all day to do a “what-if”, maybe it’s time to look for a better tool for your business.

I’ve heard plenty of companies say that finding an “80% solution fit” for their planning/scheduling will be good enough.  My experience tells me that those systems probably will fail in some way; planners especially are looking for a 100% fit or they’ll find some other way to do their job.  So it’s important that the software be capable of modeling the production process accurately, and be able to incorporate new data and create schedules quickly.  From there, planners should adjust the schedule only when “unplanned” events happen that require their review and intervention.

There is a whole spectrum of complexity to manufacturing, and the up-front effort required to implement a useful APS system should be understood before embarking on any change.  I’d like to hear about your thoughts and experiences; you can find me on LinkedIn at http://www.linkedin.com/pub/peter-nelson/5/539/52 or follow me on Twitter at http://twitter.com/petenelson.

According to you, which is the Best Production Scheduler in the market and why?

According to you, which is the best Production Scheduler in the market and why?

Satya Sahu asked this question in the “MES – Manufacturing Execution Systems” group on www.LinkedIn.com about two months ago.  Great question, and since everyone has his or her own opinion, a wide variety of answers were posted.  Not surprisingly, many were simply from advanced planning and scheduling (APS) vendors touting their own solutions.  Though I am tempted as well, I thought it would be more helpful to consolidate a listing of possibilities, and have some links to free resources on the web for anyone that is interested. 

My list is built from the information in that discussion, and I’ve tried to find additional information where I could.  If you find any errors, please visit www.planettogether.com and give me a call, I’ll correct it right away.  I’m certain that more than a few viable alternatives are missing from this list as well – not the least of which could be “spreadsheets and your brain”.  Trying to be as unbiased as I can, I’ve listed the vendors alphabetically.

In addition to this list, you may find other lists such as this one at Capterra.com to be helpful as starting points: http://www.capterra.com/production-scheduling-software?srchid=432944&pos=3.  As always, it is vital to understand your own requirements and thoroughly investigate your options.

With more than 5,000 views, many viewers have found this APS overview video to be a helpful introduction to advanced planning and scheduling for manufacturers:

[]

“Specialty” planning and scheduling systems:

Response to “A Capacity for Action”

In the January/February issue of APICS magazine, author Dave Turbide’s article “A Capacity for Action” covers many essential points regarding the choices available for capacity planning and manufacturing scheduling.  Diagnosing whether or not the current system of planning is working for a company is the first step.  It is interesting to me how many manufacturers are still using daily production meetings, emails, maintaining spreadsheets, and making on-the-fly decisions as their capacity planning system.

However, I feel that many of the considerations Mr. Turbide identifies as obstacles with supposedly sophisticated planning systems can be overcome.  Finite scheduling, optimization, advanced planning, and supply chain planning systems have been around for decades now.  New systems are emerging, while older systems likely will have issues meeting the demands of manufacturers – they are not fast enough, flexible enough, or user-friendly enough.  These systems will eventually fail to meet expectations.

With today’s APS systems, understanding and changing the optimization parameters must be intuitive and flexible.  Anything less is a failure for that particular software package, and I believe that having to accept “vendor recommendations” seems bound to fail.  I agree that optimization alone is often inadequate for capturing all necessary factors and situations, so incorporating appropriate manual overrides quickly is critical.  Creating and comparing “what-if scenarios” can be very helpful in determining the best course of action, but the system must support doing so quickly and often collaboratively.

Modern systems can reflect schedule changes in real-time – in the right situations, that can offer valuable information to an organization about the impact of those changes.  If the manufacturer cannot react appropriately to such changes, the system needs to be set to update on a frequency that makes sense.  There is vast middle-ground between responding every second, and planning once a week.

Making a change to planning systems does require some courage and determination, but it starts with identifying what improvements are needed; evaluating what tools are available to use; and understanding the costs and benefits of the alternatives.  With due diligence and cooperation between vendor and customer, there should not be major surprises once a chosen approach or system is implemented.

APS and MRP – Oil and Water?

I was recently asked some great questions by a partner of ours, some that I’ve just taken for granted over the years of working with Advanced Planning & Scheduling systems.  The key points were:

1.       How should a manufacturer use MRP and APS effectively?  (That’s the $64 question!)

2.       What should MRP do and what should APS do?

This blog is where I get to express my opinions, so feel free to agree or disagree – I’d like to hear from you either way.  I did a quick search over the web to find others’ opinions too, and I’ll share links to those at the end of this article. 

I think it is a given that a manufacturer will be using MRP for material planning, whether that is really better than APS material planning or not.  It’s the industry standard, and I don’t see MRP going away any time soon.  It does a decent job where production is more material intensive and not so resource or capacity intensive.  In reality, a lot of manufacturers will be using MRP before they consider using an APS, so retrofitting the scheduling application to MRP is almost always going to be the way APS is implemented.

It is common practice for the information transfer between MRP and APS to be a two-way flow.  Typically, at a minimum, work orders are sent from MRP to APS, they are scheduled, and the start and end dates for each work order in the APS are uploaded back to MRP.  Seems logical, using more precise and realistic start dates to drive MRP rather than using production lead times that do not incorporate capacity variables like work shifts, variable changeovers, and alternate resource availability.

This bidirectional flow gets tricky when the following are involved:

1.       Multi-level bills of material.  How will the MRP and APS “settle down” when work is rescheduled and manage the links between the levels in the BOM structure?

2.       Released, firm, and planned orders are sent to APS.  MRP is going to replan planned orders; when is it best to firm them up?

APS systems have their own material planning logic, and although it is possible to use APS systems with those features turned off, in many cases it is much better to utilize them.  I would still say this even if it seems redundant to have two material planning processes going on.  By using material links (or pegging) in the APS, the production schedule is immediately realistic – based on both material and capacity availability.  Components and intermediates will be in sync with the items that consume them.

Without material links, MRP and APS will arrive at a stable schedule, but with some approaches this could take multiple iterations of both MRP and APS to achieve stability.  Not a good thing.

I probably shouldn’t generalize on when to firm up planned orders, but I will anyway.  Firm orders are not replanned by MRP; that is a good thing if those orders are being properly planned for in the APS.  Why not just firm up every planned order?  Because those planned orders are based on the master schedule, or item forecasts, or customer orders – none of which are very stable in most environments.  So managing all of those cancellations and re-opening of orders might not be a good idea.  It will vary from place to place, but there is probably a happy medium where those planned orders which are more “real” are the ones that should be firmed up, and the others left to be replanned by MRP.  The advantage of seeing firm and planned orders in the same schedule is that capacity bottlenecks and material shortages can be seen far in advance and appropriate actions can be taken before problems occur.

MRP “action” or “exception” messages are one last thing I’ll touch on in this article.  With APS, the messages to create or cancel a production order are still valid, and those suggesting quantity changes might be too.  But with APS, with a properly configured system and good integration logic, it is probably safe to ignore all of the “reschedule” MRP messages to move dates in or out – that’s the job of the APS.

So that’s my opinion for now, but here are some links that are informative in some way.  Most of them are not that current, but they might satisfy historical curiosity.  I don’t have any type of relationship with the owners of these articles, but I found them interesting:

1.       Although it appears to be over 10 years old, this article does a fair job of explaining APS as a concept, and how it benefits MRP users.  It’s interesting to me how many of the “Leading APS Vendors” of 1996 are gone now – most of them got assumed into ERP companies.  Also, I wouldn’t expect that too many APS vendors are getting the prices they got back in ’96. http://www.autofieldguide.com/articles/059802.html

2.       This article seems to take a critical view of both APS and MRP (with the link at the bottom of the page, “MRP – No Planning System”).  It briefly mentions supply chain planning applications, and probably should be updated to look at the next generation of APS systems available today.  http://www.inventory-management.de/inventorymanagement-418.htm

3.       This page from a consulting firm’s website does explain the benefits of APS generally while promoting their services, but it might exaggerate just slightly.  The pictures of APS transforming a facility that looks like it was hit by a hurricane into a modern automated plant seem just a bit over-the-top to me, but I understand the point. http://www.isid.hk/eng/consulting02.htm

If you have questions or thoughts about this article, please feel free to contact me.  I’m happy to learn from your experiences too.

APS: The Ability to Execute

A few weeks ago I had the opportunity to talk about the advantages of APS solutions to a group of industry experts, and I made sure to touch on some of the considerations that need to be given to implementing an Advanced Planning & Scheduling solution.  Even so, a participant had a great question about how a company’s “ability to execute” should factor into a decision on whether or not to go forward with an APS project.  There is a lot of complexity within this simple sounding idea.

At first I thought the question was around how process variability might affect the schedule, but the participant was actually more interested in implementation project management.  That’s good – I’ve written a little bit about process variability and unexpected events in my posts here anyway, and I don’t want to rehash the thousands of pages of books that cover ideas around project management.  But I do want to address this type of question here.

Basically, yes, a company’s internal knowledge and resources will have an impact on any successful project, whether for APS or anything else.  I often hear comments to the effect of “Our users aren’t very knowledgeable about computers, so we could not use an APS.”  Yet these same people are capable of keeping a plant scheduled and running with little more than their bare hands in some places!

Anybody can learn something new.  We all learn something every day, whether we like it or not.  The real issue is having a good project plan and design in place that recognizes what needs to be done, who will do it, and when and how they will accomplish their tasks.  The real tragedy in many types of software sales is that this process is not done until after a decision to purchase (or not) is made.  I’ve learned that this process needs to be done from the very beginning of considering a change – see, even I can learn.

From the very beginning, the customer and the software provider need to be on the same page about the process.  Try not to take shortcuts or surprise each other.  Start by taking a close look at the current scheduling process – something is being done now, after all.  What is working well?  What is not working?  How much of an issue are the things that aren’t working?  Let’s make sure we’re looking at the impact of production scheduling across the organization and supply chain too.  Does the solution you are looking at have relevancy in these areas? 

From here, I would take a close look at those areas identified for improvement.  What causes these issues?  How serious are they?  How often do they occur?  It’s like a visit to the scheduling doctor.  By understanding the symptoms, we can find the best cure and design the treatment plan.

Not every patient wants to get better.  So the pros and cons of each course of action need to be understood by all.  What are the costs involved with staying with the current process, in terms of time, resources, and financial costs?  How do those costs compare with the other options being considered?

This type of thorough analysis during the evaluation should lead to a plan that can be implemented, and one that saves on costs.  That makes everyone happy.

On-Time Delivery - At what cost?

Advanced Planning & Scheduling (APS) systems have long promised to automate the scheduling process and make life easier for planners.  Software companies have tended to focus on being “user-friendly” with graphical planning boards and drag-and-drop on Gantt charts, or being “easy to use” with ad-hoc reporting, configurable screens, and on and on.  But what is the financial benefit to the business if a planner is now able to drag-and-drop, where copying and pasting with spreadsheets was adequate before? 

Not long ago, too little attention was paid by both software vendors and their customers to the dollar gains possible by using an APS.  Maybe we all got distracted by the colorful pretty blocks moving around on the screen, and forgot to take a close look at how drag-and-drop will really improve the business - or if the business can be improved with better scheduling at all.

That’s all changed, thankfully.  Manufacturers are much more focused on financial results from their improvement projects, and not just pretty screens.  What company would implement a new program just because it looks nicer than their current system?  A high degree of value has to be evident for an APS project to have a chance at success.

I do think there is value in drag-and-drop techniques, in the right situation.  For example, even just updating a spreadsheet with new data can take hours, so when schedules change, it is a big issue for many manufacturers.  Time and productivity for the planner is lost, and the analysis of the changes are often inadequate, when they are completed.  Just having a link to the sources of planning data and transactions is a big step in the right direction – the planner’s time and money are saved, and sophisticated ROI analysis is not needed to see that.  Manufacturers should benefit from taking a full-time planner, giving her the tools she needs to create better schedules quickly, and applying her in-depth knowledge of the company’s processes to drive better productivity results instead of spending hours updating data and participating in meetings.

With drag-and-drop, at the very least the users have the option of having a visual board to understand timelines and resources involved for each job or operation.  In cases where resource requirements are present but not overly restrictive, a graphical drag of a timeline to re-plan upstream and downstream operations can save much of that lost productivity for planners.  Not to mention setting the start time of a process exactly where it is needed to minimize inventory requirements to supply it, and having a solid estimate of the end date for delivery to the customer.  More value starts to build from a basic beginning.

On the other hand, why not go with “optimizing” your demands and supplies through a constraint-based model, and forget about manually adjusting the schedule?  It’s tempting, though there is the possibility that true optimization tools can be too complex to implement quickly (inexpensively) enough, and even tougher to modify when business conditions change (and we know they will.)  So compromises might be made, like loading resources with orders by due-date, and manually tweaking the schedule from there.  Is there a middle-ground that delivers value to a business quickly, and takes less effort to maintain with either changing optimization parameters or requiring the user to intervene excessively?

These seem to be the main parameters that most manufacturers will consider when trying to create a better schedule (within a plant or facility, I’m excluding transportation costs or lead-times here):

1.       Due-dates (to customers or inventory)

2.       Costs (labor, equipment, inventory)

3.       Productivity (minimizing changeovers between products, for example)

4.       Customer priorities

Being able to find the right balance between these often competing factors might be helpful, before resorting to a manual adjustment of the schedule (via drag-and-drop).  Business conditions will change, so it seems very important for end-users to adjust software parameters that drive the automated capacity loading process.  That part must be intuitive, or value is lost.   The drag-and-drop function must be just as responsive and intuitive when it’s needed. 

If the impact of schedule changes can be tied to those same financial performance measures that inspired the implementation of an APS system, then there can be little doubt about the value of switching planning systems.  Customers and vendors need to work together to accomplish this, and I think that requires a special kind of mindset for all, and the responsibility starts with the vendors.  Your thoughts?

When a Ripple Becomes a Tsunami

Note: Despite my recent posts involving “warnings”, “icebergs”, “garbage”, and “pain” (and now this one!), I like to think of myself as an optimist.  Maybe I need to work in some sunshine and lollypops into my posts.  Well, here we go.

I occasionally get some time to post thoughts and questions on other forums, and there is a really interesting discussion on www.linkedin.com in the “MES - Manufacturing Execution Systems” group titled “Linking MES and factory scheduling in real-time.”  Although I started the discussion, there are numerous other industry experts that offered their thoughts.  If you are interested, I would encourage you to join that group, just as I will encourage those members to consider posting in a more public area such as www.FactoryScheduling.com.

One of the most sensitive points on this discussion centered around the idea of real-time updates.  There were views that too many updates would cause too much disruption to plans, not only in one manufacturing facility but across the supply chain.  The discussion also involved ideas that disruptions do not necessarily arise from how often plans are updated, but depend on the magnitude of the change.  Many more interesting views were shared and debated.

I think a logical question is: How does a company understand the impact of change unless that change is actually applied in some way to the existing plan, and analyzed?  How do we know if that small ripple will pass, or turn into a huge tidal wave without looking at all of the downstream workload and constraints?  This question was raised in a different form in the discussion, and I also feel that defining what is a “big problem” vs. a “small problem” without actually re-planning in some way is likely to be misleading in environments with many products and/or alternate resources.

It wasn’t explicitly stated, but I think there is an undercurrent that the main element to consider is communication.  Who gets the new information, and when do they get it?  I’m of the opinion that data collected from production should update the current plan immediately.  This gives the “planner”, whether that is a software system or its human managers, the visibility needed to make the right decisions.

Armed with new knowledge, the plan can be updated.  Modern APS systems allow for unlimited “what-if” scenarios for comparison, and include automated algorithms that can create new plans in a matter of seconds.  Some can even re-plan based on event triggers, or a schedule or frequency of their own.  Even many older systems have a basic capability to keep some portion of the near-term schedule stable so that materials and resources can be staged appropriately, or lock in specific production schedules.

It’s important that the planner has the ability and time to react.  Care and time needs to be taken in most cases before a new plan is distributed internally or externally, because that is where the disruption really happens.  To me, the issue is not with inbound data updating the plan; more consideration of the outbound updates reflecting the new conditions is needed however.  Finding the right balance between reacting and planning is a challenge, but I feel that systems should allow their users the freedom to decide for themselves what works for their business.

Videos are a great way to start to visualize what can happen, and if you search www.youtube.com for “real time scheduling”, I’m sure you’ll find some thought-provoking materials.

Now that I’m done with this post, I’m going to go out and enjoy the sunshine.  I’m might even have a lollypop while I’m at it.  There, I did it!