Here's some easy supply chain math to get us started today. How many dock doors do you need to support an inbound operation that averages 100 deliveries a day, if each door can handle 20 per day? The easy answer is 5. The real answer is much more complicated than that. The real answer is: it depends. The factors on which it depends are truly endless, and touch on literally every aspect of your supply chain. To name a few:
In a perfect world, suppliers ship everything on the day and time that works best for your logistics network. In reality, they have other customers, operating hours, and production constraints that prevent the perfect scenario. These problems gets magnified in an unmanaged network, where suppliers control when freight is picked up, and when it “should” arrive, bringing us to…
Carriers make money by driving trucks, not waiting around to get unloaded. This one is really chicken-and-egg, because when things go according to plan, carriers will get unloaded as soon as they arrive. In many cases, all it takes is one truck being an hour behind schedule to mess up the rest of the day’s plan. At the same time, early-arriving carriers who aren’t keen on waiting until the planned delivery time can have a similar effect.
In my experience, the piece-price incentives driving purchasing decisions are one of the best examples of unintended consequences I’ve ever seen. This can be anything from 10% off if you order in truckload quantities to end-of-year closeouts, to traditional volume discounts. The consequences of these decisions are seen when a part used every day arrives once every four weeks in full truckload quantity.
Given the constraints of the purchasing agreements made with suppliers, material planners still have some impact on adding to the variation. The best logistics design in the world still falls apart if the parts are ordered at a different cadence than the design suggests. Also, the process of splitting a purchase order into multiple releases is often manual. With all the firefighting and expediting planners experience, who has the time?
Every computer program has to have a default, and in my experience, every MRP developer has chosen Mondays as the default due date on parts. It makes good, logical sense. Start of the production week, everyone’s ready to hit the ground running, etc. The problem-if a purchase order hasn’t been adjusted in some way by a material planner (which is also how it’s supposed to work a majority of the time), it will deliver to meet the default due date. Repeat that scenario a few hundred times, and your system has defaulted itself into a spike of activity every week.
It reminds me of that Seinfeld rental car rant about taking versus holding a reservation. In this case, anyone can unload a truck, it’s really about receiving the material on the truck. The ability of your receiving team to keep up with the flow of materials is critical to avoiding bottlenecks in the process
Batch production is awesome from a strictly production standpoint (I’m going to get grilled for that comment). High efficiency, low cycle times, etc. It’s just not awesome for anyone else in the supply chain. The impact of producing in batches is magnified back through the supply chain in a bullwhip that can create tons of havoc.
Have you ever walked out on a receiving dock and not understood why everyone was reading blog posts on their smart phones while all the dock doors were empty? Ever seen ten trucks lined up around the block waiting to get unloaded and a dock overflowing with freight? I’m guessing the items listed above helped contribute to these problems.
Life is full of variation, and a lot of it is beyond our control. The good news is in our supply chains, we have the ability to control a lot of that variation. Managing suppliers and carriers with disciplined supply chain design and processes is a great start. Installing incentives for purchasing that focus on total cost of fulfillment is also a key. Having material planners who are able to dedicate a majority of time to identifying and removing variation, and giving them the tools (systems) to do so is critical. It’s amazing how much of the firefighting disappears when the variation is reduced. Finally, receiving and production processes should be built to handle smaller batches at a higher frequency.
Obviously, a lot of these solutions are much easier said than done. The first step in beating a variation problem is admitting you have it and identifying the causes. Now it’s your turn. What contributions to (un)level flow have I left off the list?
Written by Randy Siever, Director of Operations at LeanCor
Posted by LeanCor Supply Chain Group
LeanCor Supply Chain Group is a trusted supply chain partner that specializes in lean principles to deliver operational improvement. LeanCor’s three integrated divisions – LeanCor Training and Education, LeanCor Consulting, and LeanCor Logistics – help organizations eliminate waste, drive down costs, and build a culture of continuous improvement.Facebook LinkedIn Twitter Google+