A Decision Intelligence shortlist should not begin with vendor demos. Providers are arriving from planning, optimization, visibility, risk, networks, enterprise suites, orchestration, and AI-native architectures, and without a category framework a shortlist can become a collection of impressive but fundamentally different products.

Inventory the decisions that are slow, fragmented, poorly contextualized, or overly dependent on manual coordination. For each one, document frequency, consequence, time horizon, required data, systems touched, decision rights, and the action that follows. This creates a demand-side map before the supply-side market is introduced. The physical logistics intelligence-layer discussion also explains why superficially similar providers can occupy very different architectural positions depending on where they sense, interpret, decide, and act.

The eligibility test is decision impact. Generic BI, horizontal AI, reporting, and transactional systems do not qualify by default; the technology must materially improve a real supply-chain decision That gate prevents the market from expanding into every product with a dashboard, copilot, optimization engine, or AI claim.

A useful structure distinguishes planning and optimization-led; visibility, event, and risk-led; decision-orchestration and AI-native; and suite, network, or enterprise-led approaches. These approaches can all create value, but they often solve different decision problems and operate at different levels of depth and reach.

Decision depth asks whether the platform interprets context, models tradeoffs, and changes the quality of the decision. Operating reach asks how broadly that capability spans functions, workflows, systems, partners, and time horizons. Buyers can add governance, execution connectivity, evidence quality, and implementation fit as additional criteria.

Once the structure is clear, evaluate decision fit, decision depth, operating reach, context quality, scenario and tradeoff capability, workflow and execution connectivity, governance, explainability, evidence quality, and referenceable outcomes. A large suite may be attractive when enterprise reach and installed-base integration dominate; a specialist may be stronger where one high-value decision requires deeper intelligence. The framework makes those tradeoffs explicit.

A disciplined market structure is therefore not academic taxonomy. It reduces evaluation noise, prevents false comparisons, and makes the shortlist defensible before vendor marketing begins to shape the requirements.

Market structure should follow the decisions buyers need to improve

A useful shortlist starts by inventorying the decisions that are currently slow, fragmented, poorly contextualized, or dependent on manual coordination. Planning tradeoffs, disruption response, inventory allocation, logistics exceptions, supplier risk, and cross-functional balancing may all require different forms of intelligence and different time horizons.

Only after those decision domains are explicit should buyers compare provider types. That prevents a broad suite, a planning specialist, an event-intelligence platform, and an AI-native orchestration layer from being treated as interchangeable simply because each uses similar language. The category framework should make the operating model visible before the vendor list is allowed to dominate the evaluation.

Translate the market structure into a decision inventory

Before scheduling provider demonstrations, buyers should document a small set of decision classes that matter economically: for example, responding to a logistics disruption, reallocating constrained inventory, balancing service against cost, interpreting supplier risk, or coordinating a cross-functional response to changing demand. For each decision, record the data required, the time horizon, the people or systems with authority, the actions that follow, and the cost of delay or error. That inventory turns an abstract software category into a practical evaluation model and makes it much easier to see which provider archetypes belong on the shortlist. The 2026 Market Map is designed to help organizations understand the structure of the Decision Intelligence market, evaluate provider differences, and identify the capabilities most relevant to their operating environment. If your organization is evaluating Decision Intelligence platforms or clarifying where decision intelligence fits within the broader technology architecture, I would be glad to provide the Market Map brochure and discuss the evaluation questions and provider differences most relevant to your requirements.

Request the Decision Intelligence Market Map Brochure

For technology providers

Providers may request the brochure, discuss the research framework, or contact me to confirm how their capabilities are represented in the market assessment.

Discuss the research or confirm your profile

For the past few years, transportation executives have had to manage through one disruption after another: excess capacity, collapsing rates, driver shortages, rising fuel costs, fraud, cargo theft, shifting regulations, and increasingly demanding customers.

What struck me during the transportation management discussion I moderated at the Descartes Innovation Forum was how quickly another issue has moved to the center of the conversation: technology, and particularly AI, is becoming part of the operating model rather than a separate innovation initiative.

The freight market itself remains complicated. Demand remains relatively soft, yet capacity has tightened significantly as trucking companies and drivers have exited the system. That creates an unusual dynamic in which rates can increase without a corresponding demand surge. It also changes the shipper conversation from simply negotiating lower rates to ensuring access to dependable capacity and managing greater pricing uncertainty.

For shippers, the response increasingly starts upstream. Better forecasting, inventory optimization, dedicated transportation, and network planning can reduce exposure to the spot market and prevent costly expedites. The objective is to make an emergency become an inconvenience. That is a useful way to think about transportation management because some of the largest transportation costs are created long before anyone tenders a load.

Technology is also becoming an increasingly important tool for protecting margins. One example discussed involved a repetitive process performed approximately 750,000 times each month. When converted into labor, the activity consumes roughly 1,500 employee hours every day. Automating work at that scale is not a marginal productivity improvement. It changes the economics of the operation.

Agentic AI was therefore not discussed as something sitting five years over the horizon. The conversation included active use cases involving workflow automation, voice agents, email automation, decision support, and software development. The challenge increasingly becomes deciding what should be automated, where humans should remain in the loop, and how quickly organizations can absorb the rate of technological change.

One of the most striking examples involved a proprietary transportation management system containing more than 30 million lines of code and accumulated over approximately 20 years of development and acquisitions. A 12-person team was given six weeks to recreate the system using AI-native development methods and reportedly replicated the core system in that period.

Whether every organization can reproduce that result is almost beside the point. The more important message is that assumptions about software development timelines, technical debt, and what constitutes a realistic transformation project may need to be reconsidered.

At the same time, the discussion was hardly techno-utopian. Fraud, cargo theft, cybersecurity, insurance exposure, driver qualification, and litigation all figured prominently. Transportation may be becoming more automated, but the consequences of a bad decision remain very physical.

That tension may define the next stage of transportation technology.

AI can increasingly do the work. The harder questions will be deciding which work we want it to do, which decisions still require human judgment, and how quickly our organizations can adapt to what is suddenly possible.

The post Transportation Leaders See AI Moving From Experiment to Operating Model at the Descartes Innovation Forum appeared first on Logistics Viewpoints.

The AI discussion in logistics is quickly moving beyond copilots, chatbots, and better search. At the Descartes Innovation Forum today, Descartes demonstrated something considerably more interesting: agents working across multiple logistics applications and organizations to execute parts of a representative end-to-end supply chain process.

That distinction matters.

Descartes described one of the persistent problems in logistics as the “swivel chair tax.” A shipment may move physically from origin to destination, but the information required to manage it still moves through transportation systems, compliance applications, carrier portals, email inboxes, messaging platforms, spreadsheets, and people. The applications themselves are often perfectly capable. The problem is everything that happens between them.

Descartes’ emerging answer is what it calls its Agent Control Plane. In the keynote demonstration, an order moved through four representative companies and 15 Descartes products, with 28 individual process steps occurring behind the scenes. Agents performed tasks ranging from compliance checks and information gathering to disruption detection and replanning, while the underlying Descartes applications continued to perform the operational work they were built to do.

But the most important part of the demonstration may have been what the agents did not do. At several points, humans remained responsible for consequential decisions. When a compliance issue appeared, the agent gathered the relevant information and escalated the decision. Later, when a disruption created an opportunity to rebook transportation at an 8% savings, the alternative carrier carried a trust score of just 32. The agent surfaced the option; the human rejected it.

That is a much more realistic model of logistics automation than the idea of simply turning operations over to autonomous AI. Descartes summarized the philosophy particularly well: autonomy is a dial, not a switch.

This is where agentic AI starts becoming operationally interesting. Consider what normally happens when an international shipment is disrupted. Someone discovers the change, someone determines which shipments are affected, and other people begin checking capacity, appointments, customer commitments, carrier options, and downstream consequences. Emails and phone calls start moving among multiple organizations, and by the time the problem is fully understood, hours may have passed.

In the Descartes demonstration, agents detected the disruption, evaluated its downstream impact, investigated alternatives, and coordinated information across the participating companies. Instead of simply alerting the shipper that something had gone wrong, the system could potentially deliver something much more valuable: the problem and the proposed resolution together. That represents a meaningful change in the role of supply chain software.

For decades, enterprise applications have largely waited for people to operate them. Agentic systems introduce the possibility that applications can increasingly initiate work themselves—within defined permissions and with humans inserted at the appropriate decision points. Descartes also emphasized that this is not merely a future concept. The company said it has already executed approximately 3.25 million agent operations and is opening an early-access program for the Agent Control Plane.

Just as important, Descartes is building governance around the model. Agents have identities, actions are logged and attributable, and activity can be reviewed and replayed. That may ultimately prove as important as the AI itself because enterprises will want different levels of human oversight depending on the decision, risk, and business context. They may be very willing to let agents do the investigative work, coordinate routine activities, react to predefined conditions, and bring humans the relatively small number of decisions that actually require judgment.

That was my biggest takeaway from the keynote.

The next generation of logistics automation may not be about removing humans from the process.

It may be about removing humans from all the work they never needed to be doing in the first place.

The post Descartes Takes Agentic AI Into the Logistics Workflow at the Descartes Innovation Forum appeared first on Logistics Viewpoints.