How a Logistics Company Uncovered Non-Compete Violations Using AI-Powered Digital Forensics

Overview

When a logistics company suspected that former drivers were secretly helping competitors by using company freight exchange accounts, it faced a serious challenge: there were strong suspicions but no evidence capable of withstanding legal scrutiny. By combining AI-driven digital forensics with knowledge graph analytics, BCNN transformed fragmented data into court-ready evidence, enabling the client to identify the perpetrators and protect its business.

The Challenge

A freight forwarding company suspected that several former drivers were violating their non-compete agreements by servicing transport contracts for competitors under aliases. The suspected scheme relied on sharing company accounts on freight exchanges, allowing competitors to fulfil contracts using the company’s own infrastructure.

Although management had strong suspicions, the available data was fragmented, inconsistent and incomplete. Without legally defensible evidence, the company could neither stop the abuse nor pursue legal action. Meanwhile, the practice continued to divert revenue to competitors and weaken the company’s market position.

The Solution

BCNN conducted a comprehensive digital forensic investigation combining three complementary analytical approaches:

• Geolocation analysis
• Behavioural pattern analysis
• Digital footprint reconstruction based on emails and IP login records

The investigation analysed an exceptionally large dataset:

• 200 drivers
• 1.2 million emails
• 500,000 IP login records
• 700,000 transport routes

Using BCNN Networks Notebook and knowledge graph technology, analysts mapped relationships between user accounts, devices, IP addresses and transport routes. Rather than relying on a single source of evidence, multiple independent signals were combined to identify behavioural patterns that would have remained invisible using traditional spreadsheet analysis.

The investigation ultimately revealed how company freight exchange accounts were being shared to facilitate transport contracts for competing businesses.

Implementation Challenges

The project presented several significant challenges.

First, the underlying data was highly fragmented and noisy, with missing login records, inconsistent identifiers and incomplete route information. Traditional analytical methods were insufficient to produce reliable conclusions.

Second, the findings needed to satisfy the evidentiary standards required for court proceedings. Every conclusion had to be fully traceable to the original source data, creating a transparent and auditable chain of evidence rather than relying on “black-box” AI outputs.

Finally, the sheer scale of the dataset made manual investigation impossible, requiring advanced automation while ensuring that experienced analysts remained responsible for validating every conclusion.

Results

The investigation successfully identified nine drivers who were sharing company freight exchange accounts to support competitors, confirming the client’s suspicions with court-grade evidence.

Key outcomes

• 9 individuals identified
• Court-ready evidence package prepared
• Evidence incorporated directly into the client’s legal claim
• Behavioural detection methodology established for future monitoring
• Improved protection against future revenue leakage

Beyond resolving the immediate case, the client gained a repeatable detection capability that can continuously monitor current driver activity and identify future violations before they result in financial losses.

Key Takeaways

This project demonstrates that AI-powered digital forensics can transform fragmented, incomplete datasets into legally defensible evidence. By combining knowledge graphs, behavioural analytics and digital footprint reconstruction, organisations can uncover sophisticated fraud schemes that remain invisible to conventional analytical techniques.