How do you streamline planning for tomorrow’s supply chain?

JohnWesterveld

Planning

What does your planning organization look like? Is it a series of silos?  Individual fiefdoms throwing requirements back and forth over the wall?  What are they spending their time doing?  Are they drudging through the tedious work of placing orders with suppliers?  Releasing orders to the factory?  How much time are the spending truly adding value to the supply chain and how much time are they spending doing simple, mechanical, yet important activities?

Accenture has just released a white paper titled “Supply chain for a new age” which you can get here.  In this paper, the author describes various components of the next generation supply chain and introduces the role of “Network Planners”.

A network planner would have ultimate responsibility for supply chain performance and would oversee all aspects of the supply chain from demand and distribution to detailed component supply for a group of products (a family or region or both).

As you can imagine, traditional supply chain tools and processes would quickly overwhelm any normal human, if they were to try to manage all aspects of the supply chain even for a relatively small set of items.  If the network planner is to be realized, as Accenture points out, we need to have a way of automating the basic tasks and allowing the planner to focus on making key decisions and managing the significant exceptions.

Those basic tasks really do need to be done; accepting new orders, releasing orders, sourcing decisions and supplier collaboration just to name a few. However, these are all things that while important, don’t need significant knowledge or strategic acumen to accomplish.  In other words these activities are things that could be handled automatically, thus freeing up your supply chain staff for the more complex, strategic decisions better suited for the human mind.

How would we go about automating these basic tasks?  Let’s think it through for a few examples;

Accepting new orders: Many companies support web based order entry systems. In many cases, promise dates made to customers are based on fixed delivery lead times and don’t reflect the true availability of the order based on the realities of the supply chain.  Imagine a system where the new orders that come in are automatically assessed and scheduled based on the capability of the entire supply chain.  Orders that are capable of shipping within 2 days of the request date are automatically accepted.  Orders that will be outside of that window are flagged as an exception and handled by a person.

Releasing Orders to manufacturing: Typically orders released to manufacturing need to clear two key hurdles; 1) do I have the raw material or components necessary to make this order? And 2) do I have the capacity needed to release this order?   An automation routine could easily identify the orders that are clear to release and automatically release those orders.  A more advanced system could also schedule the releases to prevent overloading by prebuilding where possible to make use under loaded time periods.

Sourcing decisions and alternate parts: Often there are options when it comes to where to source from.  Those options often come with price impacts, however, those impacts may be acceptable if they are not too great and if the alternative is late or missed delivery.  Automatic selection of a source based on lead time, prioritized by cost could allow you to automate that trade-off between margin and delivery – providing margin is within tolerance.  A more advanced tool introduces the concept of constrained supply.  Allowing the selection of an alternate if the primary source is overloaded….again monitoring margin and alerting if allowable thresholds have been exceeded.  The same applies for use of alternate parts.  In manufacturing, there are often acceptable substitutes that can be used if the preferred item isn’t available.  The use of these substitutes can and should be automated so that a planner doesn’t need to manually manage them.

Collaborating with suppliers: Once supplier agreements are in place, the effort of converting the day-to-day planned requisitions to physical orders to a given supplier should be automatic.  In some cases, supplier agreements have limits in terms of the volume that can be ordered.   With more advanced planning tools, this limit can be automatically respected by constraining the supply coming from that supplier.  Overload can be either rescheduled earlier, routed to another supplier automatically (see sourcing decisions above) or rescheduled later and flagged so that the planner can resolve the issue.   In today’s internet enabled world, key suppliers must be able to view orders, changes and cancellations from their customers via the customer’s supplier portal. The days of paper POs should be long behind us.  When a supplier responds and indicates that they can achieve what’s been requested, these orders should be automatically locked in.  Only when the supplier can’t achieve what’s been requested AND this results in an impact to customer orders or an unacceptably low end-item inventories should the buyer need to step in and look at other resolutions.

With any system that automates decision making, there needs to be oversight to ensure that the decisions being made align with corporate mandates.  This oversight can be basic monitoring of key metrics like margin, late revenue (orders or forecast), inventory levels and overloads / underloads or it could be more specific to the automation that is occurring.  For example, with supplier collaboration, the buyer could be alerted if the supplier rejected the entire order or if the order will be late by more than 10 days. If you were doing automated order promising, you could be alerted if the customer order was going to be more than 2 days later than the requested date.

Supply chain planning is getting more and more complex.  Yet we still primarily plan with software that was developed decades ago.  If we expect to meet the needs of supply chain today and in the future, we will need to look at new tools and new processes.  It’s time.

How do you envision managing tomorrow’s supply chain?  Comment back and let us know.

JohnWesterveld

John Westerveld is a demo architect for Kinaxis. Previously, John was a product manager responsible for supply management , supply chain risk management, application interface and framework, integration management and server. He joined Kinaxis over 7 years ago bringing with him 16 years of manufacturing experience, including 10 years in the information technology field. John is a qualified APICS instructor and is certified APICS CPIM.

More blog posts by John Westerveld

Discussions

  1. Dear John:

    What you have proposed should happen is already a done thing.

    Accepting new orders knowing actually when they can be delivered given the existing load – Done. (here order means an entire Job)

    Releasing Orders to manufacturing only when they have matured for execution – DOne. (Here order means individual activities that make a job)

    You are using the same word “Order” to mean a whole “job” as well as individual “activities” that make a job. No manufacturing unit manufactures something so simple as a single “activity” job. One must undertake several “activities” in a particular sequence (some may run in parallel others must follow one another) to accomplish a job.

    Sourcing decisions and alternate parts – The schedule one uses to commit new orders already factors in the sourcing lead time.

    Collaborating with suppliers: One has an updated time-table of when what needs to be supplied. One can share it with suppliers. Suppliers have a GREEN zone, AMBER zone and a RED zone to make deliveries. Suppliers are thus free to schedule their internal systems accordingly. Any change in supply lead times can still be factored in when the next reschedule happens so as to know the impact on all jobs in-hand. On each re-schedule the time-table is automatically revised. This time-table is available and updated until some distant time in future when the factory runs out of all jobs, literally.

    You have missed out on “Collaborating with Customers”: Like a time table for raw material supplies one also has access to an updated time table of deliveries. Delays are no more surprises that spring up on the due date.

    The schedule created is always actionable. No drag & drop, no resource leveling is ever required. Even complex scheduling decisions happen without any human intervention.

Leave a Reply