Key Challenges in Forecasting Logistics Demand
Brian Hoey - October 15, 2020
When it comes to securing capacity for outbound shipments of goods, manufacturers are in somewhat of a bind. The market has become increasingly carrier-friendly, which means that it’s more common than ever for transportation planners to struggle when trying to find the right carrier at the right price for the right shipment. Often, there’s not enough capacity to go around, and shippers wind up paying premium rates that eat into their bottom lines.Sure, it would be nice to skirt around this problem by securing needed capacity further out from the delivery date, but most orders only have a short planning window (a few days, maximum) from when they’re placed to when they need to be shipped. Of course, you could reserve capacity at your desired price point and parameters before the orders come in, but that poses some obvious risks. What the orders never materialize? What if the prices are actually going to drop? What if your shipping parameters suddenly change?
In the past, these risks may have seemed insurmountable. But with the emergence of transportation forecasting as a viable technology in the last few years has completely changed the calculus. Properly implemented and integrated with your TMS, transportation forecasting can make it possible to secure the right capacity at the right price far in advance while maintaining a modicum of flexibility. The question is: what’s the right way to go about forecasting logistics demand?
Expanding the Planning Window
Okay, transportation forecasting can help give you lane-by-land and carrier-by-carrier estimates of future freight pricing and availability—but how do you translate that into real value? First things first, you have to use those estimates to expand your window for planning out transportation logistics and securing the necessary freight. Yes, this means putting down money to ship products that may not even exist yet, for orders that haven’t been put in—but you’re doing so based on estimates derived by advanced analytics algorithms that have the ability to analyze past trends and other data more effectively than a human planner with a spreadsheet ever could. This means that you can be confident that you’re getting the right price when you lock it in.
Gaining Supply Chain Visibility
Of course, these advanced analytics algorithms are only as valuable as the data they can access. In other words, you need comprehensive visibility into your supply chain in order to truly make your logistics demand planning more proactive. This is often easier said than done, with data silos cropping up in the unlikeliest places up and down the supply chain. In turn, these silos can prevent you from understanding trends in demand just as easily as they can prevent you from assessing carrier reliability or analyzing the needs and inefficiencies in your transport network. To overcome these silos, you’ll mostly likely need to prioritize some type of supply chain integration that lets you share information with transport network partners via shared IT. This can pave the way for the kinds of real-time data flows that make it possible for you to effectively monitor carrier demand and prices to find the optimal times to secure capacity. Without it, you’ll be working with incomplete information that may lead you to the wrong conclusions.
Sensing Product Demand
So far, we’ve focused on the challenges that come from estimating the market for freight capacity far enough in advance to get the right price—but what about the other side of the equation? In order to perform this balancing act successfully, you also need to be able to assess future demand for your own products effectively. If you can’t, you’ll wind up securing capacity that you don’t need and upping your capital expenditures for no reason. This is, of course, one of the very reasons that the typical planning window for securing freight capacity is as short as it is—waiting for orders to actually come in seems like the safest thing. Unfortunately, waiting is a recipe for expensive logistics operations, at least in this market. Luckily, the same visibility-increasing measures that make transportation forecasting possible can be applied in other areas of your supply chain to empower smarter demand sensing. Rather than cautiously using past demand as your sole guidepost for future order volumes, you can use advanced predictive analytics trained on large datasets from up and down your value chain to get a picture of demand that gets clearer as time goes on. In this way, you can proactively match your demand to your logistics capacity.
Integrating Transportation Planning into S&OP
No supply chain operations happen in a vacuum. Like we saw above, you need visibility into other touchpoints in order to even get your forecasting off the ground—and that level of visibility is just the beginning. To create a really successful process, you need to find a way to integrate it into larger planning structures to ensure that all stakeholders are on the right page about strategy and tactics. For practical purposes, this means incorporating forecasting-powered transportation planning into your S&OP workflows. On some level, this just makes sense: S&OP is all about matching demand and capacity on an ongoing basis, and transportation forecasting offers you a way to get smarter about doing just that with regard to logistics planning. Still, this can present challenges for businesses that are used to doing things a certain way, since it requires adapting existing processes and getting stakeholder buy-in.
Finding the Right Software
We hope that by now the obvious benefits of forecasting logistics demand are outweighing any concerns about these challenges—but in some ways we’ve left the most significant challenge for last: the ability to generate plans based on orders that don’t exist yet simply doesn't exist in most TMS solutions. This means that, no matter how skillful you are at forecasting price fluctuations, demand, and capacity crunches, you will need the right IT to actually operationalize the results of those forecasts. Not only does your chosen solution need to give you the ability to create plans for non-existent orders, it also needs to connect easily and seamlessly with your existing TMS. It needs flexible integration possibilities and the ability to promote visibility rather than creating silos. This might sound like a tall order—but if you can find the right technology, it can pave the way for overcoming all of the other challenges on this list.