How BRIGHT Helped Econt Express Extract Business Value from Data With a Smart Data Strategy and ML
End Customers (2019)
Established in 1993, Econt has been growing and developing for almost 30 years to strengthen its position as an expert and Bulgaria’s leader in logistics.
Driven by the desire to provide continuous high quality, the company constantly develops innovative solutions for its individual and business customers in courier, postal, and cargo services.
The client needed a platform to provide real-time visibility and workloads related predictive machine learning (ML) mechanisms for all operational units.
The goal was to:
- Increase the accuracy of shipment volume estimates;
- Optimize and further increase the accuracy of already used formulas, as a subsequent, evolutionary step;
- Ensure efficient planning and dynamic shipments’ processing at increased load through ML algorithms.
Having the experience, business knowledge and the identified business requirements, Econt needed a strategic technology and process consulting partner with ML expertise and know-how to deliver the desired results.
As a trusted Econt partner, BRIGHT delivered a dynamic capacity management platform, based on ML and real-time monitoring.
The platform provides the following capabilities:
- Monitors activities in real-time with extremely high data quality;
- Allows client-facing operational units to manage their priorities;
- Facilitates the entire logistic network to make better-informed business decisions;
- Performs accurate and reliable machine learning forecasts;
- Provides proactive alarms and fully automated reporting;
- Easier online access for all stakeholders.
Econt team realized Splunk’s almost limitless technology scalability in terms of business logic enhancements.
There are a lot of challenges in such high value-added strategic solutions. The good collaboration and honest, effective communication between both teams was the key to success.
Of great importance for the successful outcomes was Econt’s careful guidance through the organization’s business processes. Thanks to their valuable and timely support, the development team could eliminate any obstacles.
Some of the most significant challenges were related to:
- Full and holistic understanding of the business;
- Comprehensive data analysis;
- Quality functional design of the platform;
- Extremely difficult for execution testing process of ML accuracy;
- Operational plan which warranties the high-quality ML results in terms of time.
Aiming at platform quality increase during the entire journey, BRIGHT developed a series of high value-added tasks that were not in the original scope.
The BRIGHT team developed an in-depth knowledge of the company’s processes in a short time. The project management and business analysis approaches were critical for project success. Requirements and specifications were supported by weekly demonstrations. Together both teams explored relevant and valuable use-case scenarios to get the most out of the technology.
BRIGHT’s ML approach in practice:
- Defining project objectives based on subject matter expertise and targets prioritization;
- Acquiring and exploring data having experience in data management and advanced feature engineering;
- Modelling the data assuring models quality;
- Deploying models based on comprehensive documentation;
- Plan operations and warrant monitoring and maintenance.
The platform exceeds expectations, proves its value-added potential and:
- Solves a significant business problem through ML models;
- Provides full visibility and transparency of business dynamics;
- Improves decision making in different organizational units;
- Provides a stable foundation for transport optimization;
- Improves resource and capacity planning.