Location Data in Logistics Operations: From Shipment Tracking to Cost Analysis

Location data in logistics businesses often exists but goes unused: vehicle tracking devices show points on a map, shipment information sits in another system, and expenses are gathered in accounting. When all three are combined, the resulting question is: "How much does a specific route, vehicle, and customer cost?" This article explains the building blocks of transforming location data into operational and cost visibility.

Data model: three entities, one timeline

The core entities are the shipment (what, from whom to whom, in how much time), the vehicle/driver (who, with which vehicle), and the location event (where, when). To connect location events to a shipment, the concept of a "task" is needed: a record of a vehicle carrying specific shipments within a specific time range. Without this connection, a point on the map cannot be associated with any shipment.

shipments(id, customer_id, origin, destination, planned_at, delivered_at, status)
trips(id, vehicle_id, driver_id, started_at, ended_at)
trip_shipments(trip_id, shipment_id, sequence)
location_events(id, vehicle_id, recorded_at, lat, lng, speed, source)
-- sample query: route of a shipment = trip_shipments → trips → location_events (within time range)

Integration: from tracking device to system

Fleet tracking providers usually offer an API or regular data feeds. The integration layer must normalize data formats, filter out duplicate records, and keep a log of each transfer. Data should not be lost when a connection drops and must be backfilled afterwards. Location frequency should be chosen according to business needs; "every second" data is costly and unnecessary for most analyses.

Operations screen: a map is not a report

A map answers the question "where is it right now?"; an operations manager's actual need is exceptions: shipments falling behind the plan, unexpected stops, and route deviations. A good operations screen centers on the exception list rather than the map; the map displays the context of the exception.

In logistics, the metric for data visualization is not "how pretty it looks", but "how many seconds it takes for a manager to make a decision."

Cost analysis: location + expense

Fuel, toll fees, maintenance, and driver costs come from accounting on a vehicle and period basis; distance and duration are derived from location data. When combined, unit costs can be calculated per trip, route, and customer. What requires attention here is that the distribution key (for example, how maintenance costs are allocated to trips) must be clearly defined and visible in the report; otherwise, debates over "incorrect calculations" will never end.

Customer visibility

Sharing shipment status with the customer (estimated arrival, delivery notification) requires a separate channel: customer portal, SMS, or email. The shared information should be the shipment status, not the live vehicle location; the latter is both unnecessary and touches upon the driver's personal data.

Personal data: tracking operations, not the driver

Although location data appears to be about the vehicle, it becomes personal data the moment it is linked to a driver. Under data protection regulations, purpose limitation, transparency, and retention periods must be applied. Practical principles: not collecting locations during off-work and off-duty hours (or separating them by vehicle), summarizing and deleting raw location data after a defined period, and ensuring drivers know what data is processed and for what purpose. These principles are the standard scope of our planning and mapping service.

Step-by-step implementation recommendation

1. Keeping shipment and task records in a single system (generates value even without location).
2. Tracking device integration and exception list.
3. Accounting expense transfer and trip-based costing.
4. Customer notifications.

Comparing plan versus actual

The most valuable use of location data is comparing the planned route and duration with the actuals. When the planned delivery sequence, expected arrival time, and estimated distance are recorded in the system, actual data is automatically compared against this plan. Deviations (extra stops, long waits, different routes) appear as exceptions on the operations screen; the reason for deviation is logged by the driver or operations with a brief explanation. Explanations accumulated over time help improve planning rules: in which regions and at what hours delays occur, or with which customers deliveries take longer.

Proof of delivery

Photos, signatures, or QR verifications captured at the moment of delivery are linked to the shipment record and stored alongside a location-time stamp. This record is used in both customer disputes and billing. If photos contain images of individuals, a personal data assessment must be carried out; in most scenarios, a photo of the package and delivery point is sufficient.

Vehicle and maintenance

Distance derived from location data is the foundation of maintenance planning: a maintenance task is automatically opened for a vehicle reaching a specific mileage. When fuel data (from tracking devices or fuel receipts) is combined with distance, vehicle-based consumption trends can be monitored; sudden deviations can be early signs of breakdowns or other issues. These analyses are vehicle-based and require a separate legal and ethical evaluation if used for driver performance appraisal.

Integration layer: sample flow

tracking provider API ──▶ aggregator (normalize, deduplicate) ──▶ location_events
                                                     │
                                                     └─▶ transfer log (time, record count, error)
accounting export ──▶ validation ──▶ vehicle_costs (vehicle, period, item, amount)
shipment + location + cost ──▶ materialized view: trip-based cost

Every arrow is a recording point; if data is missing, where it was lost can be found from these logs.

Reporting examples (fictional)

• Average delivery time by region and period comparison.
• Side-by-side view of trip cost and invoice amount by customer.
• Mileage, fuel, and maintenance cost trends by vehicle.
• Deviations and reasons for planned versus actual trip duration.

These reports are examples; actual indicators are derived from the business's decision questions.

Note for small fleets

The same structure applies to a fleet of just a few vehicles; the difference is scale. For small fleets, location captured on the driver's mobile app at the time of delivery may suffice instead of a tracking device; this both reduces costs and limits personal data processing to the moment of delivery. The system must use the same data model regardless of the data source; thus, transitioning to a tracking device as the fleet grows does not change the model.

Cy Logic's logistics management system is developed in this sequence and carries the "in development" label; we openly share the current scope during a meeting. You can view related services on the industry page.