01. FEBRUARY 2022 Sustainability with AI

Detailed insights and categorizing shipment structures

ONE LOGIC* conducted a pilot with a European logistics company that put the abilities of its  Data Product Builder One Data to the test. This led to the development and launch of One Data’s Logistics Enhancement—and impressive results. *We’ve been operating as “One Data” since 2023.

The logistics firm’s trucks are loaded nationwide, with deliveries then being made to the various central warehouses belonging to its trading partners. First, Logistics Enhancement provides a clear overview of the number of trips made by the trucks between sender and recipient sites, along with the fill rate (the ratio of the actual capacity used versus the total capacity available in terms of weight and volume) of each truck in the form of a heat map.

To make the overview even clearer, a distinction is made between the current fill rates of the trucks: Those that are almost empty are classified as “STG” (small truck load), partially filled trucks are designated as “LTL” (less than full truck load), and full or near-full trucks are classified as “FTL” (full truck load). Each trip is represented in green, yellow, or red—but Logistics Enhancement goes much further than this analysis feature alone.

Up to 40% fewer trips thanks to active bundling optimization 

Active bundling optimization—which allows users to combine multiple truckloads—is one of several functions and solutions provided by Logistics Enhancement. The parameter “TMAX”, which stands for “waiting period in days”, can also be applied here: Users can specify the TMAX to determine how many truckloads may be combined within any given time frame. The aim is to fill each truck as much as possible by combining several small shipments, generating more full truckload shipments.

Detailed insights and categorizing truck utilization rates

Having one hundred full or near-full trips between senders and recipients a month is a great outcome here. The sender knows that the goods do not have to be dispatched immediately, and a small delay has minimal impact on the replenishment flow of the recipient’s warehouses. Active bundling optimization shows the sender the truckloads that can be combined within this period, and by how much this reduces the number of trips (as a percentage).

If the TMAX is set at three days, the trips are reduced by 20%. At four days, it is 25%. TMAX cannot be increased at will; inventory levels must be preserved. Active bundling optimization finds a balance between load optimization and the requirements of the recipient.

The use of machine learning algorithms and forecasts has enabled this European logistics company to dispatch the same volume of goods with up to 40% fewer trips. At the same time, it has been able to identify trucks that would have been loaded incorrectly in advance, preventing pallets from being dispatched to the wrong recipient. The benefits are numerous: It decreases transaction and holding costs, reduces the volume of spoiled perishable goods, and reduces the number of vehicles on the road—all while cutting CO₂ emissions.