Back To Projects
CASE STUDY

Location Safety Analytics for Insurance: Reliance Pink Star Case Study

A unified safety score for solo female travelers across India -built with Atom Network to deliver trustworthy, data-driven location safety intelligence.

InsuranceIndustryConsumer TravelIndustryLocation IntelligenceIndustryData PipelinesCapabilityAnalytics DashboardCapabilityWorkflow AutomationCapability
Reliance General Insurance logo for Reliance General Insurance -Pink Star Safety Rating case study
Reliance General Insurance
Atom Network logo for Reliance General Insurance -Pink Star Safety Rating case study
Atom Network
Executive Summarylocation safety analytics for insurance

Reliance General Insurance and Atom Network needed a practical way to help solo female travelers assess location safety before choosing where to stay, dine, or spend time. MythyaVerse built the engineering foundation for Pink Star, a unified safety rating system that combines official crime data, nearby infrastructure, map signals, Google ratings, and user-submitted platform ratings. The result is a web-based safety intelligence experience with data ingestion, scoring, analytics, and a dashboard layer, giving travelers a clearer signal while allowing the partners to keep improving the model as new data becomes available.

Business Impact

Pink Star turned scattered safety information into a clearer decision support layer for travelers. The platform gave Reliance and Atom a data-driven product surface while preserving room for future tuning as more user and location data becomes available.

1 / 2

Dedicated Video Page

Reliance Pink Star Safety Rating case study video

A focused look at the Pink Star safety rating system built for Reliance General Insurance and Atom Network.

Watch Video

Overview

Reliance General Insurance set out to create a trusted safety rating system to help solo female travelers make more informed decisions about where to stay, dine, and spend time in major cities. The vision was a single unified safety score that blends diverse data sources into a clear, easy-to-understand signal of how safe a location feels in real life.

Problem

Safety decisions for solo female travelers often rely on fragmented signals, informal advice, or single-source ratings that miss local context.

Constraints

  • Blend official, commercial, open-map, and crowdsourced signals without presenting any one source as complete.
  • Make a complex multidimensional score easy for travelers to understand quickly.
  • Support ongoing user submissions so the model can improve as fresh experience data arrives.
  • Give business stakeholders visibility into coverage, score distribution, and contribution patterns.

Solution Architecture

  • Data ingestion pipelines normalize crime records, map layers, Google signals, and platform ratings.
  • A proprietary aggregation layer combines weighted signals into one location-level safety score.
  • Backend APIs expose location intelligence to the user interface and analytics dashboard.
  • Dashboard views help Reliance and Atom review coverage, score trends, and crowdsourced inputs.

The Challenge

Safety is multidimensional and contextual, especially for solo female travelers. Reliance needed a scoring system that captures the reality on the ground and reflects both quantitative and lived experience data. Traditional indicators like crime records or infrastructure coverage alone do not capture the full picture. Users needed a single score that combined:

  • Crime statistics
  • Proximity to healthcare, police, and transport
  • Local infrastructure quality
  • Google user ratings
  • Location based reviews
  • Crowdsourced ratings on the platform itself

Objectives

  • Build a unified safety score that blends multiple independent data sources
  • Deliver high coverage across major cities and frequently visited areas
  • Support solo female travelers with a clear, trustworthy signal
  • Enable crowdsourcing so user ratings continuously improve the model
  • Provide a dashboard for RGI and Atom to monitor patterns and insights
  • Ensure the algorithm remains adaptable as new data flows in

Our Approach

MythyaVerse worked closely with Atom Network to translate the vision into a production-ready platform. The work began with a discovery phase to map available datasets, understand constraints across regions, and design the architecture for a unified score. We developed a proprietary weighting and aggregation engine that interprets heterogeneous signals. MythyaVerse handled the engineering of the pipelines, backend services, data retrieval, storage, and the analytics layer. Atom Network led UI direction and product communication with Reliance.

Solution Delivered

  • Unified Safety Scoring Engine -A scoring model that integrates diverse data sources, processes them through a proprietary combining algorithm, and outputs a single, understandable value for each location.
  • Data Ingestion and Processing Pipelines -Automated retrieval and normalization of Google Maps data, OpenStreetMap layers, official crime records, Google ratings, and platform ratings submitted directly by users.
  • Location Intelligence Backend -Backend API services with secure database layers, user authentication, and continuous data aggregation.
  • Analytics Dashboard -A dashboard for Reliance and Atom providing coverage metrics, score distribution, crime and infrastructure breakdown, and crowdsourced contribution patterns.
  • Web Based User Interface -A clean, accessible interface that allows travelers to quickly view safety ratings for locations they plan to visit.
  • Self-Improving Algorithm -As more users submit ratings and experiences, the model gradually improves accuracy and becomes more reflective of real conditions.

Technologies Used

NextJS technology logoNextJS
Supabase technology logoSupabase
Google Maps technology logoGoogle Maps
Google Translation technology logoGoogle Translation
OpenStreetMap technology logoOpenStreetMap

Related MythyaVerse Services

Build something similar

Build a location intelligence platform

Talk to MythyaVerse about a production-grade AI, automation, or XR system with the right constraints designed from day one.

Explore More Projects