AI Supply Chain Resilience: Speak at ARC Forum 2026

A lot of supply chain teams are buying “AI” right now—and still running the same playbook when things go sideways.

If that sounds harsh, I mean it to be. The last few years taught operations leaders a blunt lesson: resilience isn’t a document, it’s a system. And systems only get resilient when you can sense problems early, decide fast, and execute cleanly across transportation, warehouses, suppliers, and customers.

That’s why the ARC Industry Forum 2026 call for speakers matters. Not because it’s another conference announcement, but because it’s an open slot for practitioners to show what actually works in AI-driven supply chain resilience and intelligent logistics operations—the kind built on real data, real constraints, and real accountability.

Why “intelligent operations” is really an AI problem

Intelligent operations means decisions get better as complexity rises. If your process falls apart the moment demand spikes, a port slows down, or a carrier rejects tenders, it isn’t intelligent—it’s brittle.

AI earns its keep when it handles the messy middle:

• Too many variables for humans to juggle (inventory, capacity, labor, ETAs, service levels)
• Too little time to decide (same-day re-plans, mid-shift labor moves, live load changes)
• Too many downstream tradeoffs (cost vs. speed vs. emissions vs. customer promises)

In transportation and logistics, the most practical AI is not a sci-fi brain. It’s a set of models and decision services that help you answer questions like:

• What’s most likely to be late—and why?
• What’s the lowest-risk plan for the next 72 hours?
• Which orders should we expedite, and which should we hold?
• Where will a supplier miss commit dates next week?

When a forum asks for talks on AI, resilience, and intelligence, it’s basically asking: Show us how you made better decisions under pressure.

The myth to kill on stage

Most companies get this wrong: they treat AI as a feature inside one tool. Then they wonder why the organization doesn’t get “smarter.”

The better framing is this:

AI in supply chain resilience is a design choice: connect sensing, decisioning, and execution—or accept that plans will always trail reality.

A strong ARC Forum session is one that demonstrates that connection end-to-end.

What resilient supply chains look like in 2026 (and what they don’t)

Resilience in 2026 looks less like “backup suppliers” and more like “rapid reconfiguration.” Dual sourcing still matters, but it’s not sufficient when disruption is multi-factor: geopolitics, weather, cyber incidents, labor constraints, and shifting customer expectations.

What I’m seeing work is a pragmatic stack of capabilities:

1) Risk sensing that’s tied to operational decisions

The point of risk sensing is action, not dashboards. If a risk alert can’t trigger a policy, a workflow, or a plan change, it becomes background noise.

Examples of decision-tied risk sensing:

• Shipment delay predictions that automatically adjust dock schedules and labor plans
• Supplier OTIF deterioration that triggers reallocation rules across DCs
• Tariff or regulatory changes that trigger lane re-costing and sourcing scenarios

2) Scenario planning that doesn’t take three weeks

Scenario planning has to run at the tempo of the business. In peak season (yes, right now in December), “we’ll analyze it next month” is the same as “we’ll eat the expedite cost.”

Modern scenario planning for AI in transportation and logistics typically includes:

• Rolling 13-week capacity outlooks (carriers + labor + warehouse throughput)
• Weekly network “what if” runs (port diversion, mode shifts, lead time variance)
• Service-level stress tests (what breaks first, and how much buffer you need)

3) Execution systems that learn

If the plan and the execution system disagree, execution wins. The only way that becomes a strength instead of chaos is when your systems learn from actual outcomes.

That means capturing and feeding back:

• Appointment adherence vs. planned
• Detention and dwell drivers
• Picking and packing cycle time distributions (not averages)
• Carrier acceptance behavior by lane and time

This is where AI forecasting, supplier performance analytics, and operational optimization stop being buzzwords and start becoming your resilience engine.

What ARC Forum 2026 is actually asking speakers to bring

The source article calls for real-world case studies, practical lessons, and strategic frameworks—vendor-neutral and designed for senior operations and supply chain leaders.

That’s a high bar, and it’s the right one.

Here’s what tends to land well with this audience:

Case studies with measurable operational outcomes

People don’t need another “we implemented a platform” story. They need specifics:

• Before/after metrics (OTIF, tender acceptance, expedite rate, forecast error, inventory turns)
• The time-to-value (90 days, 6 months, 12 months)
• What broke in pilot—and what you changed

If you can quantify impacts, even in ranges, your talk becomes memorable.

Frameworks leaders can reuse

Executives want reusable mental models. A few frameworks that consistently resonate in AI in supply chain & procurement conversations:

• Sense → Decide → Execute → Learn (closed-loop operations)
• Decision rights map (what’s automated, what’s assisted, what’s human-only)
• Data readiness ladder (from messy master data to trusted, decision-grade data)

Lessons from organizational reality

In logistics, the hard part is rarely the math. It’s the people and process:

• Changing planner behavior
• Aligning procurement, transportation, and warehouse KPIs
• Handling “shadow spreadsheets” without starting a war
• Setting governance so AI outputs don’t get ignored

A strong speaker doesn’t pretend these issues don’t exist—they show how they handled them.

Three session angles that will stand out (and drive real leads)

If your goal is to get selected and spark serious conversations afterward, pick an angle where peers can copy your approach. Here are three that fit the Forum themes and the AI in Transportation & Logistics campaign focus.

1) AI forecasting that actually improves resilience

The most useful demand and supply forecasting sessions are the ones that admit the truth: forecast accuracy isn’t the goal—decision quality is.

A talk that will get attention:

• How you linked forecast outputs to inventory policies (safety stock, reorder points)
• How you handled promo spikes and new product introductions
• How you monitored model drift and seasonality changes
• The “human override” rules that prevented bad automation

What makes it lead-friendly: everyone in the room knows forecasting pain is expensive, and they’ll want to compare notes.

2) Transportation AI: from ETA predictions to automated re-planning

Predicted ETAs are table stakes now. What leaders want next is decision automation:

• When to re-route
• When to split shipments
• When to switch modes
• When to renegotiate service commitments

A compelling case study here includes:

• Your exception strategy (which exceptions matter, which don’t)
• The triggers and thresholds you set
• How dispatchers/planners were trained to trust (and challenge) recommendations

Memorable line for your abstract:

“An ETA prediction is only valuable if it changes what you do next.”

3) Supplier risk + procurement intelligence that doesn’t live in spreadsheets

Procurement teams are drowning in scorecards that don’t connect to operations.

A standout session shows how AI models (and good analytics) connect supplier signals to:

• Allocation decisions
• Expedite approvals
• Contract terms (lead time bands, variability clauses)
• Dual sourcing triggers

And it gets even better if you include geopolitics and regulatory volatility, because that’s one of the Forum’s highlighted themes.

How to write a proposal ARC will accept (150–250 words that don’t waste anyone’s time)

A winning proposal reads like a mini case study, not a mission statement. Aim for clarity and proof.

Use this structure:

1. The operational problem (one sentence)
2. The constraint (data gaps, legacy systems, peak volumes, limited headcount)
3. What you built/changed (process + tech, not tech alone)
4. How AI was used (forecasting, optimization, anomaly detection, NLP for docs, etc.)
5. Measured results (even partial)
6. What attendees can copy (3 takeaways)

Example takeaways that feel “executive-level”

• How to set decision thresholds that prevent automation from creating noise
• A KPI tree that aligns transportation, warehouse, and procurement actions
• A governance model for who can override AI recommendations—and when

If you’re vendor-side, keep it educational and concrete. The fastest way to get rejected is to pitch a product demo in disguise.

People also ask: what makes an AI logistics talk credible?

Credible AI in logistics talks are specific about data, decisions, and adoption. You don’t need to share proprietary details, but you do need to show you’ve been in the arena.

Include:

• The data sources you used (TMS events, WMS scans, EDI, telematics, order data)
• The decision points improved (planning, tendering, slotting, replenishment)
• The failure modes (bad master data, drift, missing events, user distrust)
• The operational controls (override rules, escalation paths, monitoring)

If you can say “here’s what we’d do differently next time,” you’ll instantly separate yourself from marketing-only speakers.

If you’re trying to drive change, speaking is a shortcut

Sharing a real AI-driven resilience story does two things at once: it helps the community, and it creates a forcing function inside your company. You’ll tighten your metrics, clarify your architecture, and get crisp about what worked.

If you’re part of our AI in Supply Chain & Procurement series, this is the moment to turn your internal learning into external leadership—especially heading into 2026 when boards are demanding risk visibility, and customers are demanding reliability.

The ARC Industry Forum 2026 call for speakers is looking for practical strategies and frameworks that leaders can apply immediately. If you’ve built something that improved supply chain resilience, reduced exceptions, improved service, or made procurement and logistics decisions faster, you’ve got a story worth telling.

What’s the one operational decision in your network that would improve overnight if your team trusted the data—and acted on it automatically?