AI Exchange: Inside the Last Mile—AI, Delivery Engagement, and the New Standard for On-Time and In-Full

Date: April 30, 2026; Time: 11:00 a.m. ET; Live Webinar: 1 hour. What if your routes could learn from every single delivery? In this AI Exchange session, we examine how moving beyond static service time assumptions unlocks a new level of fleet performance. Traditional routing treats every stop as predictable, but in reality, each one is shaped by order size, product mix, field conditions, dwell requirements, and team readiness. AI and machine learning transform the model by learning from actual delivery behavior and continuously applying that intelligence to future routes. From a supply chain perspective, last-mile delivery accounts for 40–53% of total logistics costs and, with e-commerce growth, has become the most expensive segment of the supply chain. Descartes Systems Group competes against Manhattan Associates, Blue Yonder, Oracle, and SAP in route planning and real-time visibility, offering the Descartes Routing & Mobile, Descartes MacroPoint, and Descartes ShipRush product suite.

At the same time, performance planning does not stop there. Real-time communication, accurate ETAs, proactive updates, and shared visibility keep shipments, drivers, and customers in sync, helping teams stay coordinated and reducing unnecessary follow-up calls. From a supply chain perspective, OTIF (On-Time In-Full) is the primary performance metric that retailers enforce on suppliers, and a threshold of 95%+ is the industry standard. Walmart, Target, Costco, Kroger, and Home Depot impose 1–3% penalties for OTIF violations in their vendor agreements. Real-time ETA is calculated by combining traffic data, weather, service time history, driver experience, and vehicle telemetry data.

The speakers are Cyndi Brandt, Vice President of Fleet Solutions at Descartes, and Sergio Torres, Senior Vice President of Product Management at Descartes. From a supply chain perspective, VRP (Vehicle Routing Problem) and VRPTW (Vehicle Routing Problem with Time Windows) algorithms are transitioning from classical OR (Operations Research) techniques to machine learning–based approaches. Reinforcement learning is being used to learn from driver behavior, traffic patterns, and service time deviations. NVIDIA cuOpt, a GPU-accelerated VRP solver, can deliver results 100x faster than traditional CPU-based solvers. Descartes Routing serves major customers such as Cisco, Sysco, Pepsi, Coca-Cola, Republic Services, and Waste Management.

From a supply chain perspective, how last-mile visibility translates into customer engagement directly affects NPS (Net Promoter Score) and CSAT (Customer Satisfaction) scores. Delivery window narrowing through SMS, email, push notification, and in-app messaging channels reduces customer complaints stemming from delivery time violations. FedEx, UPS, DHL, and Amazon Logistics are the players raising the standards of the last-mile experience. Research from Boston Consulting Group (BCG) and McKinsey shows that more than 60% of consumers prioritize delivery experience in their purchase decisions. Ultimately, Descartes' AI Exchange webinar clearly outlines the AI-powered new standards of last-mile routing and customer engagement.

Key Takeaways:
\n1. Descartes webinar will be held on April 30, 2026.
\n2. Static service time is being replaced by real delivery behavior.
\n3. ML learns from order size/field/team dynamics.
\n4. Real-time ETA ensures alignment between shipment, driver, and customer.
\n5. Cyndi Brandt and Sergio Torres are the speakers.

[43756]