2020 puts supply chain planning at the top of the agenda in 2021
Logistics planning expert Robert Oliver tells us how supply chain planning is evolving and how you can master the process to drive better results
Supply chain planners are in a tough spot following on from a hugely disruptive period in 2020. That’s one of the conclusions of our new report, The Retail Supply Chain & Logistics Planning Report 2021, which is free to download now.
In rapidly transforming environments, many are struggling to keep up with the sudden changes and emerging bottlenecks in their supply chains. The report found that 33% of respondents reported a significant gap between their plans and their eventual execution, hampered by issues around technology performance, integration, visibility and a lack of flexibility in operations and forecasting capacity.
They need help, but that help is increasingly now at hand in the form of new technology and processes that can drive big gains in logistics operations’ efficiency and accuracy.
We spoke to an expert in the field, Robert Oliver, an Industry Process Consultant at report co-author Dassault Systèmes, to get the inside track on how to drive meaningful change in the planning process in this Q&A.
[This interview has been edited and condensed]
The issues in the planning process
Alex Hadwick, Editor, Reuters Events Supply Chain: Let’s start with what's going wrong in supply chains in terms of the logistics planning. What are the main gaps in people's abilities and what are they struggling to do at the moment?
Robert Oliver, Industry Process Consultant, Dassault Systèmes: The biggest challenges that we hear are problems with forecasting. That really became clear in 2020, when forecasts were out the window pretty much. Forecasting, visibility of movement of products through the supply chain, and keeping costs down, are probably the top three challenges that we see in the supply chain. Then below that may be reacting to change, and certainly 2020 saw a lot of change.
Alex: How do you think organisations are handling the problem of existing in a time of pretty much unprecedented change and therefore historical data diminishing in its usefulness?
Robert: In terms of making better forecasts, the answer is to put more weight on recent forecasts and on collaborative forecasts.
You need a forecasting system that isn't just looking at historical data and making a statistical forecast. You also need a collaborative process where the statistical forecast is reviewed by people who are domain experts … so that the statistical forecast can be amended by people's knowledge of what's actually going on that may not be supported by history.
Alex: Why are we not further along in terms of instituting route planning technology as an industry?
Robert: That's one of the things that turned up in the survey. Almost 30% of respondents were doing manual planning, which is kind of crazy. In the supply chain planning business, one of the major forces is manual planning, typically done in Excel. That's not going to produce optimal plans. It's going to take a lot of work to produce any plan, because you're depending on a lot of knowledge in people's heads.
The system would produce infeasible plans and the planners would spend a lot of time essentially fixing the plan, and they just abandon the system. They go back to Excel
Why are people doing that to such a large degree? I think part of it is upfront cost. You have to implement a system. Now, that system should easily pay for itself, after you have it implemented, but you have to have that upfront investment and be willing to do that and convince management that it makes sense.
Another aspect of that is there are plenty of people that have route planning systems or transportation management systems, but it doesn't produce a really good plan. The system did not take into account all of their constraints or their business goals, it was an off-the-shelf piece of software. Usually, you can't do a ton of configuration in that. The end result is that the system would produce infeasible plans and the planners would spend a lot of time essentially fixing the plan, and they just abandon the system. They go back to Excel.
Helping planners through deploying the right tech
Alex: Let's talk a little bit about the planning process. What kind of technological aids can help those subject matter experts? What would you advise companies to have in place as a supporting mechanism to be able to react rapidly to changing environments?
Robert: It depends on the time horizon that you're looking at and it depends on where in the supply chain you're looking. There's different tools for each area.
If you're looking at strategic planning, what do I want to do a year from now? Two years from now? Do I need to build a new distribution centre or warehouse? Do I even want to consolidate distribution centres in a geography? For that kind of long term planning, which is related to capex and that you can't change quickly, then you really want a tool, some sort of a system that allows you to look at your whole supply chain, and evaluate ‘what if?’ scenarios. What if we did consolidate these distribution centres? Or what if we did build another manufacturing facility and it added this capacity for this type of production. That's for what we call the strategic horizon.
You need to have a forecasting tool that allows you to rely on more than just statistics
When you when you get down the executional level, you still want tools that can do ‘what if?’ scenarios, but now it becomes more important to get granular and not just look at things in buckets of time of months or quarters or years.
You have to make use of what you have, so again, you need tools that can do 'what if' scenarios based on your forecasts and, again, you need to have a forecasting tool that allows you to rely on more than just statistics. Then it’s down to operational execution.
Let's say you're talking directly about logistics and you are routing trucks. There you really need something that can put together a feasible plan quickly and replan and react to changes, because now things are happening very quickly.
If the system doesn't understand all of the constraints that you have, then it will produce a plan but it won't be feasible and then someone has to fix it or customer service goes down
Trucks are on the road, you've got inbound shipments coming into a Distribution Centre (DC) and you've also got outbound trucks may be going to stores or customers and you need to be able to adapt to change. If a supplier suddenly has a load for you to pick up and bring back to the DC, you want to be able to dynamically plan that to an appropriate driver to go pick that up and, and make that backhaul back into the DC. So, you need an awful lot of capabilities down at that operational level, from rapid planning to making feasible plans.
If the system doesn't understand all of the constraints that you have, then it will produce a plan but it won't be feasible and then someone has to fix it or customer service goes down.
An example of that would be sending a large truck out to a small residence in a in a neighbourhood with a truck is not going to easily fit in the streets, or sending a truck to a customer location that does not have a loading dock, or where the truck should have had a lift gate on it to get the product out of the truck and you send a trailer that has no lift gate. That is not a feasible plan.
You either catch that and fix the plan, or you don't catch it.
So, on the logistics and operational side of thing, you really need a very capable system that can be configured to understand all of your constraints, and put together really optimised low-cost plans rapidly and change them rapidly.
Alex: A lot of that also comes down to visibility and affecting meaningful change in a plan relies on multiple levels of visibility. With the disruption going on currently, noticeably in international shipping, is the lesson to focus on generating more visibility higher in the supply chain than we do right now?
Robert: Obviously, visibility at all levels is important. It becomes harder higher in the supply chain, especially when you're talking about international shipping , but you really need to couple that with the ability to adjust to what's happening. So, even if you don't have some Nirvana of visibility, once you do get information, A) can, can you see the ripple effects? And B) can you react to them and adjust your plan?
You really want systems that understand the relationship between the different parts of the supply chain and propagate that information
So you really want systems that understand the relationship between the different parts of the supply chain and propagate that information. So, if you find out that a particular shipment is late, what does that mean downstream? What are the ripple effects? Is that going to cause a production issue? Is that going to make something else late? Is that going to change your delivery plan because the product is not going to be there to deliver on time? Those downstream ripple effects and understanding those and reacting to those, I think, are at least as important as the visibility.
Alex: Let’s finish by talking about digital twins. I wondered how you feel about what role they might play in the near future?
Robert: We love digital twin at Dassault Systèmes. That's what the company is all about. There's different kinds of digital twin in the warehouse space, and manufacturing space.
I see it as really is two kinds of digital twins. There's the 3d visual twin and then the logical digital twin that comes down more to the planning level, at the operational timeframe.
For example, we have all sorts of cool technology for doing 3D simulations of a factory or warehouse before you build it, where if you're going to move operations around, you can do simulations and 3D visualisations and ask will this equipment fit in an existing building? Etc.
Then when you get into more of the supply chain and logistics space, the digital twin tends to morph into modelling your supply chain and all of your constraints, your capacities, your business goals, and being able to produce feasible plans.
Because you've got this digital twin of your available truck resources and capacities and drivers, you're able to come up with feasible plans that meet your business goals, whether that is on-time delivery or lower costs, as well as react to whatever happens and deal with that.
Conclusion
Business leaders need to start thinking about how to use new analytics technology not just to analyse and understand the past, but to make better decisions for the future. Those that use big data to feed big business decisions will create an optimal supply chain network infrastructure. But keep in mind that technology will only enable change if it is suited to your specific strategy.