- Case study ↓
- AI/ML
- Computer Vision
- Video Analytics
Car Wash Analytics: Case Study
Use the cameras already on site to record vehicle check-ins, wash times and service stages, with footage operators can review.
The demo
See Car Wash Analytics at work
- Industry
- Car wash operations
- Input
- Existing car wash camera footage
- Output
- Session dashboard, Processed video, API access
- Built with
- Computer Vision · Video Analytics · Dashboard · API
The problem
Manual logs leave gaps in the wash story
Paper logs and staff estimates can miss vehicles, wash times and service steps. POS records show a sale, but not what happened in the bay or how long each wash took.
The system turns existing camera footage into checkable session records. Each time or service status can be reviewed against the video that produced it.
The solution
What Car Wash Analytics does
Vehicle and Equipment DetectionClassifies the vehicle type and detects wash equipment such as the washing head, each with a confidence score.
Vehicle Check-In and Check-OutEntry and exit of the wash area are stamped automatically, with the total wash time calculated.
Service Time TrackingWash duration per session, comparable across bays and shifts to find where time is lost.
Service and Cleaning StatusEach service stage, such as water, soaping, washing, external or internal wash, marked detected or not for every session.
Multi-Station and Multi-Bay ManagementCameras, bays and locations organised under one shared account.
Recorded and Live Video InputsStart with recorded footage for an assessment; connect live streams once the deployment is ready.
Processed Video ExportFootage with the system’s detections drawn over it, for review.
Secure Cloud Storage and API AccessFootage and results kept in a secured cloud account, with an API to pass results to other software. Scoped per deployment.
Technology
How it’s built
Every step, from the existing car wash camera footage to the results.
Step 1
Video input
- Existing cameras
- Recorded or live stream
Step 2
Detection and tracking
- Vehicle type
- Wash equipment
- Bay movement
Step 3
Session analysis
- Check-in and check-out
- Wash duration
- Service stages
Step 4
Operator output
- Session dashboard
- Processed video
- API access
Who it’s for
- Independent Car Wash OperatorsUnderstand throughput, busy periods and wash durations at a single site.
- Multi-Site OperatorsCompare stations and locations on the same measurements, with one reporting approach.
- Car Wash Software and Equipment ProvidersExplore adding vehicle and service analytics to an existing product through an API.
Case study
Car Wash Analytics: the full case study
- Overview
An AI-powered car wash monitoring system that uses computer vision to watch vehicles and cleaning activity through IP cameras. It turns video footage into structured service records, so operators can see how vehicles move through the site and how long each service takes.
- Problem
Car washes often rely on manual supervision to monitor vehicles, track cleaning activities and record service times. That makes it hard to spot delays, measure individual service stages and keep consistent records across several service bays.
- Solution
ML Bench developed a computer vision solution that detects and tracks vehicles, classifies them as SUV, sedan, full-size SUV or motorcycle, and recognises people, cleaning equipment and cleaning activities. It records each vehicle's entry and exit and times every service stage.
YOLO11s handles vehicle and object detection, object tracking keeps each vehicle's identity across frames, and temporal activity recognition picks out cleaning activities from video sequences. Live camera streams arrive over RTSP, results are served through a FastAPI backend, and the system is designed for edge deployment on NVIDIA Jetson hardware with TensorRT.
- Results
- Better visibility into service activities across every bay
- Potential delays identified from service stage timings
- Less reliance on manual video monitoring
- Data-driven decisions on service efficiency and resource use
Have Questions?
Find the right solution
for now
Usually not. The system reads the CCTV-style feed a site already has, from a live stream or recorded video. We assess camera angles and coverage for each site, and a sample video shows whether the current setup is enough.
Why ML Bench
Why teams choose ML Bench
Built for real-world footageWe build computer vision for real cameras and scans: retail, sports, healthcare, logistics and safety.
Starts with the data you haveExisting cameras, recordings or image archives are usually enough for a first proof of concept.
Custom to your setupWe adapt the model to your camera angles, image quality, classes and workflow, not the other way round.
One team, pilot to productionData, model training, dashboard, API and deployment handled by the same team.
Next AI/ML project →
T-Shirt Counting Matching
AI system that detects, tracks and counts T-shirts during printing, counting each shirt once.
Free assessment
See What Your Cameras Can Tell You
Send us a short video from one of your car-wash bays. ML Bench can assess what the footage makes possible and identify a suitable next step.
- UI/UX Design
- Web Development
- Mobile Apps
- SEO
- AI/ML Solutions
