From Tetris to Puzzles

A secondary transport network distributing product – fuels and chemicals – to more than 2,000 retail points of sale – dealers – spanning 2 continents. Important efficiency problems in terms of unit costs – cost per liter – due to a fixed, basic process that is triggered upon dealers’ conveniences and their particular short-term needs and objectives, resulting in variable order sizes impacting cost, below-benchmark fleet utilization rates, and an above-benchmark average freight cost.

A process that had been designed in the first place as demand-responsive to satisfy the customer was behaving more like a Tetris, with lots of wholes, resulting in decreasing margins for both sides. How to “puzzle” the pieces together and avoid having “wholes”?

 

The ambition

In essence, it is to disrupt the cost by moving from a demand-driven operating model based on point of sale replenishment to a supply-driven one, based on supply process optimization or, in other words, to take control of the operation from the supply side in order to design and run a process capable of reverting all the endemic inefficiencies, minimizing unit costs – cost per liter delivered –, increasing margins for the transport company, and also offer better dealer margins.

A win-win approach that enables a smoother negotiation of new terms with the dealer network.

 

Addressing it

Make of the whole supply process a service and bundle it to customer as a key element of the customer value proposition.

Internally, this is implemented as a service unit in the form of two regional transport excellence hubs that act as a shared service center operating as a business unit.

Process-wise, the shared service center acts as a transport digital twin that breaks down the cost structure into three core subprocesses: demand, scheduling, and transport, and identifies within each of them all the key levers that are necessary to tackle in order to maximize load sizes and routes.

An important of those levers is the development of an automatic stock replenishment tool that includes inventory control at the point of sale and considers the whole inventory needs of the full network for proper demand planning. This enables targeting full-load trips to multiple delivery points and maximizing the number of trips per truck. It also involves more deliveries per point of sale in a short period of time and therefore more flexibility for peak needs or contingencies, which, in turn, feed the learning process of the planning algorithm.

Other actions include the development of policies such as freight incentives based on optimal dealer order sizes, variable pricing based on conditions such as frequency, time, and size, fleet management, improvement of loading and waiting times, etc.

 

Creating impact

95% of full load, same-size trucks being dispatched.

Variable instead of fixed service levels, with flexible delivery hours.

Weekly planning of resource needs.

From 1 service shift to 2-3 shifts according to customers’ operating hours. Scheduling aligned with point of sale on a weekly basis.

27% of total cost improvement, coming from: 18% due to load optimization, 9% due to optimization of truck utilization rates.