Introduction
Supply chain visibility has become a critical capability for businesses operating across international suppliers, carriers, ports, warehouses and distribution networks. Companies increasingly know where their shipments are, when they are expected to arrive and whether something has disrupted the journey. But that information does not automatically solve the problem.
A delayed truck can still miss its warehouse appointment. A late container can still leave a factory short of materials. An alert can reach a logistics team without anyone having enough time or information to respond.
That gap between knowing and acting is becoming one of the biggest challenges in modern logistics.
In an October 21, 2026 article, project44 examined how Unilever is moving from shipment tracking toward what it calls decision intelligence. The case study described how the consumer goods company uses transportation data to support dock scheduling, exception management and inventory decisions. project44 reported that Unilever reduced detention and demurrage costs by 20% by using transportation data to trigger decisions rather than simply displaying shipment status. project44
The broader lesson is relevant well beyond one multinational company. As global supply chains become more exposed to tariffs, geopolitical disruptions and changing trade routes, businesses need more than a dashboard showing where goods are. They need systems and processes that help people decide what should happen next.
Background and Context: Why Traditional Supply Chain Visibility Is No Longer Enough
For years, logistics teams relied on phone calls, emails, spreadsheets and carrier updates to track shipments.
Finding out where a truck was could involve contacting a driver, checking with a carrier or waiting for a warehouse update. Ocean freight could be even harder to monitor because a shipment might pass through multiple carriers, ports and transport providers before reaching its final destination.
Digital visibility platforms changed that process.
They brought shipment events, estimated arrival times, carrier updates and transportation information into a central view. Instead of chasing individual updates, teams could monitor shipments through a shared system.
That was a major improvement. But visibility has a limitation: it tells a company what is happening, not necessarily what to do about it.
Consider a truck scheduled to arrive at a warehouse at 10:00 a.m. Its estimated arrival time changes to 11:30 a.m.
A traditional visibility system reports the delay. A more advanced operating system can help answer several additional questions:
- Should the warehouse appointment be changed?
- Is the receiving dock still available?
- Will the delay affect production or inventory?
- Does the carrier need to be contacted?
- Should another shipment be prioritized?
- Does the customer need an updated delivery commitment?
These are operational decisions, not tracking events.
That distinction is at the center of the move toward decision intelligence.
Latest Update: How Unilever Is Moving From Visibility to Actionable AI
project44’s October 21 article, written by Tom Zyer, described a discussion at Reuters Events: Supply Chain Europe 2026 between Carlo Andrea Carmenati, Primary and Inbound Operations Excellence Lead Europe at Unilever, and project44’s Tom Lintsen.
The discussion outlined five lessons from Unilever’s efforts to connect real-time transportation information with operational decisions. project44
1. Stop spending time chasing shipment status
Before real-time visibility, logistics teams often spent a significant part of their day answering basic questions about shipment locations and delivery status.
Unilever’s Carmenati described a working environment in which his team used to receive around 100 calls, emails and team messages each day asking where trucks and orders were.
With real-time visibility, that information became available through a central system.
The change is more important than it might initially appear.
When employees spend less time chasing updates, they have more time to investigate exceptions, coordinate with carriers and resolve problems before they affect customers.
The objective is not simply to make shipment tracking more convenient. It is to change how logistics teams spend their working hours.
2. Turn changing arrival times into operational decisions
One of the clearest examples in project44’s case study involves dynamic dock scheduling.
Imagine a truck that is expected to arrive at a warehouse at 10:00 a.m. The latest estimated arrival time changes to 11:30 a.m.
A basic tracking system displays the revised time. But the warehouse may still have the original appointment in its schedule.
That creates a risk of waiting time, congestion, missed slots and additional costs.
Unilever is testing a process in which a significant ETA change can trigger a proposed adjustment to the truck’s dock appointment.
The workflow described by project44 includes three stages:
- The system identifies a meaningful change in the truck’s estimated arrival time.
- A new dock appointment is proposed.
- A warehouse employee confirms the adjustment, allowing the schedule to be updated.
The human confirmation step is important. AI can identify a change and recommend a response, but warehouse operators still provide the context needed to confirm whether the new appointment makes operational sense.
This is a practical example of AI supporting a decision rather than independently controlling every part of the process.
3. Use agentic AI to investigate transportation exceptions
Transportation delays often create additional administrative work.
A missing shipment update may require a team member to contact a carrier, request information, follow up and manually enter the response into a transportation management system.
When hundreds or thousands of shipments are moving through a network, those small tasks can consume substantial operational time.
The Unilever discussion described an agentic AI approach in which an AI agent can send emails or make calls to request missing shipment information, retrieve the response and enter relevant data into the transportation management system.
The important difference is that the system does more than create another notification.
It can help gather the missing information needed to understand an exception.
That changes the workflow from:
Alert received → Employee investigates → Carrier contacted → Information collected → System updated
to a more assisted process:
Exception identified → AI gathers information → Context is organized → Employee handles decisions that require judgment
This does not mean every exception can be resolved automatically. Some situations require negotiation, operational knowledge or direct human intervention.
The opportunity is to reduce the manual work surrounding routine exceptions so employees can focus on the problems that genuinely need their attention.
4. Connect visibility to measurable financial outcomes
The strongest technology projects demonstrate operational value rather than simply producing more data.
In project44’s account, Unilever reported a 20% reduction in detention and demurrage costs through improved transportation data quality and accountability.
Detention and demurrage are related but distinct cost categories.
- Detention generally refers to charges associated with keeping a carrier’s equipment beyond the permitted time.
- Demurrage generally refers to charges for keeping containers at a terminal or facility beyond the allowed free time.
Accurate arrival and departure information can help businesses establish when shipments reached and left a location.
Geofenced timestamps provide a more objective record than relying entirely on manual updates.
Better information can also support inventory planning. If a company understands actual transit times more accurately, planners can make better decisions about safety stock and potential delivery risks.
project44’s case study also connects shipment visibility with on-time, in-full performance, or OTIF. By identifying route deviations, extended stops and at-risk deliveries earlier, logistics teams may have more time to intervene before a delay affects a customer.
The 20% reduction is a result reported in project44’s account of Unilever’s experience. It should not be interpreted as a guaranteed saving for every company adopting similar technology.
5. Connect the entire transportation journey
Global supply chains rarely depend on one carrier or one transportation mode.
A product may move from a supplier by truck, travel by ocean freight, pass through a port, move to another warehouse and eventually reach a distribution center.
A delay in one part of that journey can affect every stage that follows.
Unilever works with approximately 150 carriers, according to project44’s article. Those carriers do not all connect through the same technology. Some use APIs, while others require different identification information or mobile applications.
This makes network connectivity an ongoing operational challenge.
The next opportunity is to connect information across modes, not simply within individual shipments.
For example, a delayed ocean vessel could affect:
- The truck scheduled to collect the container.
- The warehouse receiving appointment.
- The availability of materials at a production facility.
- Inventory allocation.
- Customer delivery commitments.
When ocean and road transportation data are connected, companies can better understand how a delay in one mode affects the rest of the journey.
That is the broader ambition behind end-to-end decision intelligence.
Expert Insights and Analysis: Visibility Is the Foundation, Decision Speed Is the Advantage
The Unilever example points to a larger change in how businesses should evaluate logistics technology.
For years, companies measured visibility by how many shipments they could track, how frequently their systems received updates and how accurately they could estimate arrival times.
Those capabilities remain important. But they are increasingly becoming the starting point rather than the final objective.
The next question is how quickly the organization can respond to the information.
The difference between visibility and decision intelligence
| Supply chain visibility | Decision intelligence |
|---|---|
| Shows shipment location | Helps determine what action is needed |
| Displays estimated arrival times | Connects ETA changes to dock and delivery decisions |
| Generates disruption alerts | Helps investigate and prioritize exceptions |
| Tracks carrier performance | Supports decisions about carrier and route alternatives |
| Provides shipment history | Helps improve inventory and operational planning |
Visibility provides awareness. Decision intelligence attempts to turn that awareness into a response.
That response might be a recommendation, a workflow change or an automated action, depending on the level of risk and the organization’s operating rules.
AI needs trustworthy data and operational context
An AI agent cannot reliably solve a logistics problem if the information it receives is incomplete or inaccurate.
For example, an ETA change might indicate a genuine delay. It might also result from an outdated carrier update or an incorrect location event.
If an AI system reacts to unreliable information, it can recommend the wrong dock appointment, contact the wrong party or create unnecessary work.
Companies need to establish data quality standards, identify trusted sources and monitor how their systems behave under different operating conditions.
This is especially important when multiple carriers and transportation modes are involved.
Automation should not remove human accountability
The Unilever example retains human confirmation for the dock scheduling process described by project44.
That is a sensible approach for decisions that can affect warehouse operations and downstream commitments.
AI can identify patterns, collect missing information and propose responses. People should remain involved where judgment, safety, customer commitments or financial consequences require approval.
A successful AI implementation does not have to eliminate human decisions. It should help people make those decisions with better information and less delay.
Broader Implications: What Actionable AI Means for Global Supply Chains
The shift from visibility to decision intelligence has consequences across industries.
1. Retailers could improve inventory availability
Retail businesses depend on reliable transportation to replenish stores and distribution centers.
A late container or truck can affect product availability, particularly when inventory levels are already low.
If logistics systems can identify delays earlier and connect them to inventory information, planners may have more time to transfer stock, change replenishment priorities or communicate revised delivery dates.
The value is not limited to tracking a shipment. It comes from understanding which products, locations and customer commitments are affected.
2. Manufacturers could reduce production disruptions
Factories often depend on the timely arrival of components and raw materials.
A delayed shipment can create a production bottleneck even when the rest of the factory is operating normally.
Connected transportation and inventory information can help manufacturers identify which inbound shipments are critical and which delays require immediate attention.
This can support better production planning and more targeted safety stock decisions.
3. Logistics providers could reduce manual exception handling
Carriers, freight forwarders and third-party logistics providers manage large volumes of shipment updates.
When information is missing or inconsistent, employees may spend considerable time chasing confirmations.
AI agents could assist with repetitive communication, data collection and system updates.
That could allow logistics teams to focus more on customer service, complex exceptions and network planning.
4. Global companies need faster responses to tariff changes
The project44 article published on October 13, 2026, titled Redesigning around tariffs? Decision speed is what makes it work, explored how companies are changing suppliers, sourcing locations and distribution networks in response to trade policy.
It described changing freight patterns on Asia-to-US trade lanes, including a growing share of shipments originating from Thailand, Vietnam and Cambodia relative to China.
The article also introduced the idea of decision lag, meaning the time between a change in trade conditions and the point when a company’s network actually responds.
A company might identify a new supplier or alternative manufacturing location, but the change can take time to implement. Carrier capacity, warehouse arrangements, inventory positioning and supplier onboarding all need to follow.
That is why supply chain visibility is increasingly connected to strategic decision-making, not just shipment tracking. project44
5. Geopolitical disruption makes network intelligence more important
project44’s September 25 article, Weaponized trade routes are now a strategy, examined how geopolitical tensions can disrupt major shipping corridors.
The article discussed the Red Sea and Strait of Hormuz as examples of how trade routes can become exposed to political and security risks.
It also argued that companies must prepare for disruptions that overlap rather than treating every event as an isolated incident.
A business might face tariff changes, port congestion, longer shipping routes and carrier capacity constraints at the same time.
In that environment, a dashboard that reports individual shipment delays is not enough. Companies need to understand which disruptions affect their network, which alternatives remain available and which decisions should be prioritized.
The ability to connect information across carriers, modes and locations becomes a strategic capability. project44
For more coverage of supply chain technology, AI-driven logistics and global trade, explore The Tech Marketer’s supply chain technology coverage.
Related History: From Shipment Tracking to Intelligent Logistics Networks
The evolution of supply chain technology has moved through several stages.
First came manual tracking. Logistics teams relied on phone calls, emails, carrier paperwork and spreadsheets to understand where shipments were.
Then came digital visibility. GPS, electronic data interchange, carrier APIs and cloud-based platforms made shipment information easier to collect and share.
Next came predictive analytics. Businesses began using historical and real-time data to estimate arrival times, identify potential delays and understand transportation patterns.
Now comes decision intelligence. AI systems are increasingly being designed to connect information with recommendations and operational workflows.
This does not mean earlier technologies are becoming obsolete. Decision intelligence depends on reliable tracking, connected systems and accurate data.
The difference is that the objective has expanded. Instead of only knowing what is happening, companies want to know what action will produce the best outcome.
What Happens Next: Five Priorities for Supply Chain Leaders
The move toward actionable AI will require more than purchasing another visibility platform.
Companies should focus on five priorities.
1. Measure the cost of manual exception management.
Identify how much employee time is spent chasing shipment updates, contacting carriers, correcting data and coordinating delayed deliveries.
This creates a baseline for evaluating whether automation is actually improving operations.
2. Connect shipment data with warehouse and inventory systems.
An ETA is more valuable when it can be connected to dock availability, inventory requirements and customer commitments.
Companies should identify where information remains disconnected.
3. Start with repeatable, low-risk workflows.
Routine tasks such as collecting missing shipment information or organizing exception records may be suitable starting points for AI assistance.
More sensitive decisions should retain appropriate human approval.
4. Build visibility across transportation modes.
A company should understand how ocean, road, rail and air freight interact across the full journey.
A delay in one mode should not remain isolated from the teams responsible for the next stage.
5. Measure outcomes, not software activity.
Track detention and demurrage costs, on-time delivery, OTIF performance, exception resolution time, inventory accuracy and manual workload.
These measures provide a clearer view of whether decision intelligence is producing meaningful business value.
The companies that make progress will not necessarily be those that automate the most tasks. They will be those that identify where faster and better decisions can protect customers, inventory and operating margins.
Conclusion: The Future of Supply Chains Is About Acting on Information
The next stage of supply chain visibility will be defined by what companies do with the information they collect.
Unilever’s experience, as described by project44, shows how real-time shipment data can support dock scheduling, exception investigation and better transportation decisions. The reported 20% reduction in detention and demurrage costs offers a concrete example of how better information can translate into financial outcomes.
But the larger lesson is not about one platform or one company.
Global supply chains are becoming more complicated. Tariffs can change sourcing strategies. Geopolitical events can disrupt shipping corridors. Port delays can affect road transportation, inventory availability and customer commitments.
Companies need systems that help them understand those connections and respond before a problem spreads.
Visibility tells a business where its goods are. Decision intelligence helps it determine what to do next.
That difference could become one of the defining competitive advantages in global logistics.
Frequently Asked Questions
1. What is supply chain visibility?
Supply chain visibility is the ability to monitor products, materials and shipments as they move through suppliers, carriers, ports, warehouses and distribution networks. It helps businesses understand shipment locations, estimated arrival times and potential disruptions.
2. How is AI improving supply chain visibility?
AI can help analyze shipment data, identify unusual events, investigate exceptions, recommend operational changes and automate routine communication. It can help turn tracking information into actions that support logistics operations.
3. What is decision intelligence in supply chain management?
Decision intelligence is the process of using data, analytics and AI to support or automate operational decisions. In logistics, this can include changing dock appointments, prioritizing delayed shipments and identifying alternative transportation options.
4. How did Unilever reduce detention and demurrage costs?
According to project44’s October 21, 2026 case study, Unilever reduced detention and demurrage costs by 20% through improved transportation data quality and accountability, including the use of geofenced arrival and departure timestamps.
5. What is agentic AI in logistics?
Agentic AI refers to AI systems that can carry out multi-step tasks toward a defined objective. In logistics, an agent might request missing shipment information, collect a response and update a transportation management system, subject to the company’s controls.
6. Why is real-time shipment tracking important?
Real-time tracking helps businesses identify delays and disruptions earlier. When connected to warehouse, inventory and transportation workflows, it can give teams more time to respond before problems affect customers or production.
7. How do tariffs affect supply chain visibility?
Tariffs can change sourcing locations, shipment volumes, transportation routes and inventory strategies. Businesses need visibility across suppliers and freight networks to understand how trade policy changes affect costs and delivery plans.
8. What is the difference between supply chain visibility and supply chain resilience?
Visibility helps a business understand what is happening across its supply chain. Resilience is its ability to prepare for, respond to and recover from disruptions. Visibility supports resilience by providing information needed to make timely decisions.
Sources & References
1. project44, “From Visibility to Actionable AI: How Unilever Is Building Certainty Into Global Logistics”
October 21, 2026.
https://www.project44.com/blog/from-visibility-to-actionable-ai-how-unilever-is-building-certainty-into-global-logistics/
2. project44, “Redesigning around tariffs? Decision speed is what makes it work”
October 13, 2026.
https://www.project44.com/blog/redesigning-around-tariffs-decision-speed-is-what-makes-it-work/
3. project44, “Weaponized trade routes are now a strategy. Here’s what that means for your supply chain.”
September 25, 2026.
https://www.project44.com/blog/weaponized-trade-routes-are-now-a-strategy-heres-what-that-means-for-your-supply-chain/
4. project44, Content Hub
https://www.project44.com/content-hub/
5. NIST, “AI Risk Management Framework”
https://www.nist.gov/itl/ai-risk-management-framework
6. GS1, Global Traceability Standard
https://www.gs1.org/standards/traceabilityproject44+3
