Transforming critical operations at regional scale

The problem

Transandina ran six terminals between Puerto Cortés and San Salvador on three different management systems, inherited from successive acquisitions. The consequence was not an IT problem but an operational one: nobody could say where a vehicle was without picking up the phone, dwell time was estimated at month end, and the network's declared capacity did not match what it actually achieved on any given day.

The board had budgeted a yard expansion at two terminals. The original brief was to appraise that investment.

The solution

The capacity study showed the terminals were not saturated: they were unbalanced. Two were running at 94 % while two ran at 51 %, and allocation was decided by commercial habit rather than by load. The recommendation was to postpone the works and fix visibility first.

Access control and vehicle identification were installed at all six terminals, the three systems were brought under a single data model, and a dynamic bay allocation layer was built on top. The civil works followed, cut back to one terminal and to a third of the original budget.

Process

1. 01Network diagnosticEight weeks measuring actual dwell time by terminal, by shift and by vehicle type.
2. 02Pilot at Puerto CortésAutomatic identification and bay allocation at the busiest terminal, against a prior baseline.
3. 03ConsolidationMigration of the three databases to a common model, running in parallel for sixty days.
4. 04Regional rolloutExtension to the remaining five terminals, including the expansion works at one of them.

Technologies applied

• Automatic vehicle identification (RFID + plate OCR)
• A single data model on PostgreSQL
• Dynamic bay allocation layer
• Dashboards by role: yard, terminal and board
• Integration with the existing ERP over an API
• Redundant link between the terminals and the control centre