Project Overview
Meridian Logistics Group, a publicly listed logistics company headquartered in Port Klang with operations across 8 ASEAN countries, commissioned SCS Synergy to develop a custom real-time business analytics platform. The company manages over 50,000 shipments monthly across road, sea, and air freight, and needed a unified analytics solution to replace a fragmented reporting ecosystem comprising 15 separate Excel-based reports, three legacy BI tools, and numerous manual data collection processes.
Our team designed and built a modern data platform that ingests, processes, and visualises operational data in real time. The platform consolidates information from GPS fleet tracking, warehouse management systems, customs clearance databases, and financial ERPs into a single source of truth. Executive dashboards provide instant visibility into KPIs including on-time delivery rates, fleet utilisation, warehouse throughput, and profitability by route and client.
The analytics engine goes beyond descriptive reporting with predictive capabilities powered by machine learning. Route optimisation models analyse historical traffic patterns, weather data, and port congestion forecasts to recommend optimal shipping routes. Demand forecasting helps warehouse managers anticipate capacity requirements up to 30 days in advance, reducing costly last-minute warehouse space procurement by 35%.
We implemented role-based access controls allowing different stakeholder groups to access relevant dashboards. Operations managers monitor real-time fleet positions and delivery status. Finance teams track revenue per tonne-kilometre and cost variance. The C-suite receives automated weekly briefing reports with trend analysis and anomaly detection, highlighting areas requiring executive attention.
The Challenge
Meridian's data landscape was severely fragmented. Critical operational data resided in siloed systems that did not communicate with each other, forcing analysts to manually extract, reconcile, and compile reports. The monthly performance report required approximately 48 person-hours to produce, and by the time it reached decision-makers, the data was often two to three weeks old. This lag made it impossible to respond quickly to operational disruptions or market shifts.
Data quality was another significant hurdle. GPS tracking data from the fleet of 800 vehicles arrived in inconsistent formats depending on the tracking device manufacturer. Warehouse systems across different countries used different units of measurement, currency formats, and classification codes. Historical data spanning five years existed in various formats including CSV exports, Access databases, and paper records that had been partially digitised. The platform needed to harmonise all of this into a consistent, reliable analytical foundation.
Our Solution
SCS Synergy implemented a modern data stack built on Apache Kafka for real-time event streaming, Apache Spark for batch and stream processing, and ClickHouse as the analytical data warehouse optimised for sub-second query performance on billions of rows. The ETL pipeline normalises incoming data from 23 different sources, applying quality checks, currency conversions, and standardised classification codes before loading into the warehouse.
The frontend was built as a React web application with D3.js for custom data visualisations that go beyond standard charting capabilities. Interactive maps display real-time fleet positions with colour-coded delivery status. Sankey diagrams visualise cargo flow through the logistics network. Drill-down capabilities allow users to navigate from high-level KPIs down to individual shipment details with a single click.
For the predictive analytics module, we trained gradient boosting models on three years of historical shipment data to forecast delivery times with 91% accuracy within a 2-hour window. The route optimisation engine processes real-time traffic data from Google Maps Platform and port authority APIs to suggest alternative routes when disruptions are detected. All models are retrained weekly on fresh data to maintain accuracy as conditions evolve.