The Challenge

In the fast-paced world of telecommunications, the client needed to maintain a leadership position in the competitive 5G landscape. As the company grew, so did customer dissatisfaction and churn due to inconsistent data coverage, particularly in high-traffic business areas. To overcome these hurdles, this major telecommunications provider partnered with Entrada to optimize its network infrastructure and address two key issues: 

Inconsistent Performance and Delays: Traditional geospatial data processing tools struggled with the scale and complexity of the data, leading to inefficient operations.
Frequent SLA Violations and High Operational Costs: Inefficiencies in the data processing pipeline resulted in missed SLAs and escalated costs, impacting the bottom line and stakeholder confidence.

Pinpointing strategic locations for new 5G tower installations was critical to addressing these issues and optimizing the company’s network infrastructure. The only way to achieve this was to conduct a comprehensive analysis of customer usage patterns and cellular signal data.

The Solution

With Databricks Labs’ proprietary Mosaic, a cutting-edge distributed geospatial processing engine, alongside Apache Sedona (formerly GeoSpark) for enhanced spatial operations, Entrada developed a custom geospatial algorithm to revolutionize data processing efficiency and performance. The algorithm seamlessly ingested, transformed, and joined point and polygon datasets. 

By simplifying the scheduling, orchestration, monitoring, and alerting functions, our team streamlined the entire data processing lifecycle, enabling the client to harness the full potential of their geospatial data. Utilizing Sedona’s spatial operators allowed for efficient geospatial joins between disparate datasets, significantly reducing processing times.

The cornerstone of the solution was the generation of a heatmap, utilizing Kepler.gl, that visually represented the optimal locations for 5G tower placement. This intuitive and data-driven approach enabled decision makers to allocate resources strategically, resulting in maximizing coverage and signal strength within high demand areas.

The Algorithm: Pinpointing Tower Locations

Geospatial Joins: Mosaic’s functions effortlessly joined cellular signal points to their respective business polygons, allowing aggregation of signal metrics within business footprints.
Coverage Analysis: Business areas with high popularity but poor average signal strength and connection drop rates were flagged for further investigation.
Recommendation Engine: Results were enriched with demographic and land-use data and fed into a model that prioritized ideal cell tower sites. Kepler.gl visualizations aided in narrowing down optimal locations.

The Results

Through collaborative efforts, the client was able to revolutionize their 5G network to drive business growth, enhance customer satisfaction, and cement their position as a trailblazer in the dynamic telecommunications industry:

25% reduction in customer complaints related to poor signal quality

15% decrease in customer churn

20% increase in revenue

30% reduction in operational expenses

About Entrada
Entrada is a Databricks-focused consulting and implementation partner backed by Databricks Ventures. Entrada harnesses the power of Databricks to help customers accelerate their AI + data initiatives. Our expertise in AI/ML, Databricks, and analytics is centered around industry-centric solutions. Our mission is to simplify complex data + AI challenges and support end-to-end transformations, delivering future-ready solutions fast.