Trax Technologies CEO advocates disciplined AI adoption for smarter transportation management

At the American Supply Chain Summit, Blake Tablak, chief executive of Trax Technologies, made a pointed case for a slower, more disciplined approach to artificial intelligence in transportation management. Rather than treating AI as a shortcut to savings, he argued that the real advantage comes from better questions, cleaner data and a leadership culture that tolerates experimentation.

Tablak, who said he is not a natural early adopter, framed the issue as one of judgement rath...

er than novelty. In his view, many companies rush towards new tools before they have worked out what they are trying to learn. That, he suggested, is why so many AI initiatives stall: the technology may be powerful, but it cannot compensate for vague objectives or fragmented information.

The problem is especially acute in transport spend, where data is often scattered across freight audit providers, parcel systems, ocean carriers and regional operations. According to Trax, executives frequently struggle to build a complete picture because no single team owns the layer that brings those feeds together. Tablak said he has spoken with finance leaders frustrated by the lack of visibility, and recalled a conversation with a major automotive manufacturer in which a supply chain executive’s answer to a cost-cutting request was simply to ship less.

His broader point was that companies need to think in terms of three layers: those that generate data, those that consolidate it and those that use it to make decisions. Once that structure is in place, he said, AI can begin to answer questions that were previously too time-consuming or too complex to handle manually.

Trax illustrated that argument with a series of client examples. In one case, a large industrial manufacturer wanted to understand the trade-off between longer payment terms and carrier performance. Within minutes, the company found that extended payment periods were tied to weaker service on certain lanes, while rates were also above market levels. The result was not a lengthy tender process but a more direct discussion with existing carriers.

In another example, a consumer technology group asked how often it was shipping the same package repeatedly between the same locations. The answer, Trax said, was thousands of times a day, pointing to a structural inefficiency that was pushing consolidation costs on to retail partners. A medtech company used similar analysis to identify where shipping surcharges should apply, while another deployed a performance index to rank carriers by cost, speed, quality and emissions, depending on shifting business priorities.

That flexibility matters because the benefits of AI in logistics are already being demonstrated elsewhere. Case studies from logistics technology firms suggest that route optimisation and predictive systems can deliver measurable gains in fuel consumption, planning time and delivery performance. Appther has said a platform it built for SwiftLogistics cut fuel costs by 28% and improved delivery speed by 35% within 90 days. TruckFlow Logistics reported a 40% reduction in route-planning time and a 23% fall in fuel spend after introducing AI dispatching. Other providers have cited lower costs, faster planning and fewer breakdowns from automated routing and maintenance systems.

Those examples point to the same conclusion Tablak was making on stage: AI works best when it is embedded in a clear operational model, not bolted on as a fashion statement. He urged finance and logistics leaders to consolidate fragmented freight systems, begin with the business questions they most need answered and accept that useful outputs often emerge only after several rounds of refinement.

For Tablak, the real prize is not simply faster freight audit or cleaner reporting. It is the ability to surface answers that no one else in the organisation can provide. In his telling, leaders who want to use AI well must become relentlessly curious about the business itself, and willing to learn through trial, error and iteration.

Source: Noah Wire Services