IndiaLogistics and transport2024 – ongoing

Allcargo Logistics

Machine learning for LCL capacity, route consolidation and proof-of-delivery checks

Allcargo says AI cut daily departures by 8-10% and receivables by 15-20%

WorkingCompany's own claim

Problem

Allcargo ran about 1,500 route departures a day in its less-than-container-load (LCL) network. Capacity commitments, pricing and the checking of proof-of-delivery documents all depended on manual judgment.

Approach

The IT team under group CIO Kapil Mahajan built models on historical data to forecast demand by trade lane, including seasonal trends and global events. The forecasts fed capacity commitments and dynamic pricing. Machine learning also looked for ways to consolidate routes, and company experts validated the results. A neural-network vision model checks signatures and stamps on delivery documents. It runs on the mobile app and works offline.

Results

8-10%reduction in daily departures, from about 1,500, without hurting service levels, per AllcargoCompany's own claim[1]

15-20%reduction in outstanding receivables after automated proof-of-delivery checks, per AllcargoCompany's own claim[1]

Takeaway

The three uses each targeted a cost or cash-flow line that could be measured: departures, spot pricing and receivables. The figures are the company's own, and the article gives no start date.

Sources

With AI at the Wheel, Allcargo Steers Logistics into the Future
IBRS6 Aug 2024News article

Source summaries

[1]IBRS6 Aug 2024

The article describes three AI uses at Allcargo: LCL demand forecasting with dynamic pricing, network optimisation and computer-vision checks of proof-of-delivery documents. It relays the company's figures of 8-10% fewer daily departures and 15-20% lower outstanding receivables.

Full details are in the original sources.

Added 2 Nov 2026 · Updated 2 Nov 2026