South KoreaLogistics and transport2018 – ongoingSOCAR
AI-based fleet pricing, placement and inspection
A car-sharing company that credits AI operations for its first annual profit now runs 1,000–2,000 cars per staff member
Problem
SOCAR rents about 20,000 cars that customers pick up without staff, so damage, dirt, accidents and maintenance needs are not seen by employees at hand-over. For years the company lost money, with an operating loss of KRW 20.9 billion in 2021.
Approach
Since 2018 the company has moved operations onto in-house models. Demand forecasts drive where cars are placed and how they are priced, and a booking optimiser reallocates reservation slots to raise use per car. Its damage-detection system, Exida, reviews customer photos and alerts accident managers when repair is needed; photo analysis decides when a car needs washing, and driving data is used to flag unreported accidents and insurance fraud.
Results
Takeaway
The company attributes its 2022 turnaround to AI-based operations, but the published figures do not separate the effect of AI from other changes in the business, such as revenue growth of 37.6% that year. The company reports that one employee now handles 1,000 to 2,000 cars, up from about 50 at founding.
Categories
- Industry
- Logistics and transport
- Business function
- Production and qualityMarketing and sales
- Innovation type
- Process innovationBusiness model innovation
- Adoption
- Built in-house
- Evidence
- Company's own claim
Sources
Source summaries
Reports SOCAR's 2022 results, revenue of KRW 397.6 billion (up 37.6%) and its first annual operating profit of KRW 9.4 billion after a KRW 20.9 billion loss in 2021. The company credited AI-based operations pursued since 2018, including demand-based car placement and pricing, booking optimisation, automated detection of unreported accidents, more efficient maintenance and AI analysis of driving data for fraud.
Describes SOCAR's AI tools across a fleet of about 20,000 cars, including the Exida damage-detection system, photo-based washing decisions and weather-aware wash scheduling. It reports that one employee now manages 1,000 to 2,000 cars, against about 50 at founding, and cites company IR material showing maintenance costs down 19.1% year on year in the second quarter of 2023.
Reports a memorandum between SOCAR and the AI inspection start-up Weflo. It says SOCAR already uses AI to judge from customer photos whether a car needs repair and plans to add automatic damage photography without customer involvement. No results are reported; joint testing is listed as a goal.
Full details are in the original sources.
