Page 15 - Logistics News Oct Nov 2020

Demand Planning

other hand, the condition could be temporary, such product at each location at any time. In theory, we
as recently when we went through a significant could continuously tune our parameters every time
supply shortage of toilet paper. Though traditionally we forecast. In practice, we might not want to do
this would be a commodity product, toilet paper, that since our mind can’t grasp all these dynamics,
hand sanitizers and disinfectants recently moved and too many changes to tasks and processes
into the category of short supply, and prices moved challenge the auditability of what we are doing.
up accordingly. Certainly, when we have major changes, we will
Fixed forecasts are out. Flexibility in forecasts is want to tune or change our parameters. And we
in. And in reality, that kind of always was the truth. would want this to occur automatically. Whether
Products, and the calculating approach used to the systems are autonomous or just send alerts to
plan them, may not be fixed values with a fixed the user, we do need the system kind of running on
relationship between demand/supply and price. some level of autopilot since this is just too much for
Whether a regular part of the landscape, the normal a human to figure out.
process within the cycle of product introduction and
fulfillment, or emergency market issues, grappling Conclusions – dynamic, expanded, continuous
with these issues requires a widened horizon in and autonomous
looking at, understanding and, therefore, forecasting We know that the idea of using history alone in
and managing demand and supply. The impact of forecasting the future was somewhat questionable,
weather, a social trend, or some kind of national at times. While that approach may not be obsolete,
emergency can be understood and expressed as it is limiting, since now we can look at an expanded
part of an analytical model, i.e., as fields, parameters, view of our markets, customers, channels,
values and equations, to capture, calculate and suppliers, carriers and so on, and see the many
communicate said impact and its resultant values. demand-impacting factors as well as the dynamic
interrelationships between them that we never saw
Enter the non-fixed parameter before. And if we are customers of a large supply
So, let’s get back to toilet paper. What changed chain provider, we can now gain the benefits of an
here? If we think about traditional forecasting aggregated view of history and multiple in-process
systems, the forecasting parameters are fixed – in logistics flows to monitor and react to changes. We
the toilet paper example, we use history + some can see that by the SKU for each time increment,
safety stock value and demand flows by channel/ allowing constant tuning and optimising of many of
customer/location. Within safety stock, there is a the factors where the money gets spent and made
value that either a planner or the system set based – price, order quantities, supply volumes, inventory
on history. In this example, my field has a fixed investments, safety stock, on and on.
parameter/a fixed forecast method. This is the way We now have the ability to capture all that data,
we did things. Now, however, history, channel and have a machine digest and continuously learn from
safety may not mean much with a range of factors it, and produce some insights. Using formulas, the
influencing how we plan even the simplest things. system can pick the right forecast method based on
History can tell us a lot, but not maybe in the fixed all of the factors.
way it did before.
Grocers and distributors know, either by history Why would we want to do this?
or intuitively, the dynamics that previously impacted We now have the ability to dynamically change
actual sales or shortages. For example, severe plans, safety stock, production values, warehouse
weather warnings will create a run on food, water, stores and so on based on the variety of demand
and toilet paper. Hence, the lower limit on safety impacts and their interrelationship at a specific
may no longer apply, since demand will spike and timeline and location, giving us much better
we don’t want to run out. perceptions and actions. Based on this, we can
How would this flexibility in parameters, improve sales and profits by producing at the right
methods and data work? If I have an ability to flex time, reducing logistics costs or excesses; and at the
or change my parameter for toilet paper for each right price, optimising prices across the life cycle of
storm, say based on category of the storm, I can a product and with more dynamic pricing schemes
select a time period that contains the last major based on market characteristics.
storm of this magnitude. I can select with the sales This is increasing sales, increasing profit. A real
history date-range and leverage that forecast. I can win. But we do have to make a lot of changes to get
also change the actual way I utilise those specific to this point. And that means upgrading the data
values, for example time/data range, smoothing and and technology. If the recent past has taught us
best fit algorithms. anything, those who had a more robust and smarter
In essence, most of the fields, ranges and the supply chain operating model or had the smarts to
actual forecast method used can change for each react early are doing okay. •

October/November 2020 | Logistics News 13