This Vietnamese logistics company managed mid-mile and last-mile delivery operations across the country's three major economic regions — Ho Chi Minh City, Hanoi, and Da Nang — operating a fleet of 420 vehicles completing approximately 15,000 deliveries per day. Vietnam's rapid e-commerce growth, which had accelerated 35% year-over-year, was straining the company's manual routing processes to the breaking point.
Route planning was performed by a team of 11 dispatchers who used a combination of a basic mapping tool and deep local knowledge of Vietnam's unique traffic challenges — from motorbike-congested urban streets to seasonal flooding on rural delivery routes. Despite their expertise, internal analysis estimated the fleet was driving 32% more kilometers than necessary, resulting in excess fuel costs of approximately VND 48 billion (roughly USD 2 million) annually. On-time delivery rates had dropped to 68%, well below the 85% industry standard.
The company's biggest competitive threat came from the logistics arms of major e-commerce platforms that were investing heavily in AI-optimized delivery networks. The CEO recognized that without a step-change in operational efficiency, the company would lose its independent logistics business within two to three years.
Pertama Partners conducted an AI Readiness Audit that analyzed four months of GPS telematics data, delivery completion records, and real-time traffic patterns across all three operating regions. Our analysis revealed that the primary inefficiency was not route planning per se but the assignment of delivery batches to vehicles — dispatchers were grouping deliveries by district rather than by optimal route sequences, creating systematic backtracking.
The AI Pilot Program deployed a two-layer optimization system across 100 vehicles in Ho Chi Minh City. The strategic layer optimally assigned deliveries to vehicles based on geographic clustering, package size and type, customer delivery windows, and vehicle capacity. The tactical layer dynamically adjusted routes in real time based on traffic conditions, road closures (common during Vietnam's rainy season), and new orders received during the delivery day. The system incorporated Vietnam-specific factors including motorbike traffic density predictions, known flooding zones during monsoon months, and the narrow alley access constraints common in Vietnamese cities.
Team Training focused on transitioning dispatchers from route planners to exception managers. Executive Training helped leadership understand how to use the optimization platform's analytics for strategic decisions about fleet composition, hub placement, and service area expansion into secondary cities.
"Vietnam is one of the most complex last-mile environments in Southeast Asia. Pertama Partners built a system that understands our flooded alleys, our motorbike traffic, and our monsoon seasons better than any international logistics platform could."— Nguyen Van Minh, Chief Operating Officer
Every transformation starts with a conversation. Let's discuss your challenges and opportunities.
Discuss Your Challenge