In Part 1 of this blog series, we introduced the OSM-APLS and delta-coverage metrics to assess the accuracy of OSM distance calculations. We also evaluated the various routing cost configurations and…

Swiggy's blog serves as a platform for industry insights, covering a wide range of topics such as emerging food delivery trends, consumer preferences, market analysis, and innovations in the food service sector. Through  articles, case studies, and interviews with industry experts, Swiggy offers  knowledge and actionable strategies for restaurant owners, food entrepreneurs, and stakeholders in the food delivery ecosystem. Readers can learn about effective marketing techniques, operational best practices, customer engagement strategies, and technology advancements shaping the future of food delivery services.

Swiggy Bytes

The post discusses improving the accuracy of OSM distance computation by addressing missing road segments, missing connectivity, and incorrect one-way road designations using machine learning algorithms. It introduces map matching using a Hidden Markov Model (HMM) and details an unsupervised algorithm inspired by the CatchMe algorithm to detect and fix errors in the OSM database, followed by a self-supervised algorithm to improve detection in congested urban areas.

The OSM Distance Service Part 2: Fixing Roads