Claims processing · Production

Allstate automates vehicle damage claims review from weeks to days with super.AI

The problem

Allstate needed to scale the review of thousands of vehicle damage images and videos to reduce claims adjustment times from weeks to days — a process that was manual and time-consuming.

Workflow diagram · grounded in source
1
Images uploaded via API
trigger
“Integrated via API to upload data programmatically and fetch results automatically”
2
AI pre-processing with mask R-CNN
ai_action
“We use a mask R-CNN model to process the input images and provide a collection of masked images, each mask covering everything but one car in the image”
3
Route masked images to processors
routing
“The router decides, according to the project's quality, cost, and speed requirements, which processing sources are best to handle the task and sends it there for processing. The router can choose to use multiple sources for a single task…”
4
Human segmentation of vehicle parts
human_review
“The data program generates a data processing interface that our human data processors use to segment the image. The task instructions tell them how to segment the image, and a selection of tools make the process as painless as possible. …”
5
Damage identification by processors
human_review
“a new set of data processors can identify damaged and undamaged parts of the now segmented body and windows”
6
Final QA via combiner module
validation
“the combiner produces a trust-weighted combination of them as the final output. This is one of the most complex parts of the system. The result is a high quality segmented image output with all body and window parts of all vehicles segme…”
7
Results fetched via API
output
“Integrated via API to upload data programmatically and fetch results automatically”
Reported outcome

Allstate was able to review thousands of images and video footage of vehicle claim damage and expedite the claims adjustment process from weeks to a few days, improving customer experience and reducing costs.

Reported metrics
Claims adjustment timefrom weeks to a few days
Customer experienceimprove customer experience (CX) through reduced claims cycle times
Efficiency and costimprove efficiency while reducing costs
Reported stack
mask R-CNNAPI
Source
https://super.ai/case-studies/top-american-insurer
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

Allstate was able to review thousands of images and video footage of vehicle claim damage and expedite the claims adjustment process from weeks to a few days, improving customer experience and reducing costs.

What tools did this team use?

mask R-CNN, API.

What results were reported?

Claims adjustment time: from weeks to a few days; Customer experience: improve customer experience (CX) through reduced claims cycle times; Efficiency and cost: improve efficiency while reducing costs (source-reported, not independently verified).

How is this claims processing AI workflow structured?

Images uploaded via API → AI pre-processing with mask R-CNN → Route masked images to processors → Human segmentation of vehicle parts → Damage identification by processors → Final QA via combiner module → Results fetched via API.

Related claims processing cases