This website uses cookies so that we can provide you with the best user experience possible. Cookie information is stored in your browser and performs functions such as recognising you when you return to our website and helping our team to understand which sections of the website you find most interesting and useful.
Design and creation of an automatic quotation generation application
Design and creation of an automatic quotation generation application
Type of project
- MVP solutions (further developments in progress);
- Technical documentation;
- Digital marketing recommendations (SEO, SEA);
- Graphic charter of the bodywork;
- Body shop website;
- The solution for generating tailor-made quotes.
Context and issues
- Development of an application allowing a customer to automatically generate an estimate by taking a picture of a bodywork damage ;
- Redesign of the commercial site and the graphic identity for a better acquisition of prospects (SEO issues, adwords, customer journey).
Missions carried out
- Solution for creating quotes for a body shop based on photos of potential customers;
- Feasibility study for the use of Artificial Intelligence to solve the problem, monitoring solutions and state of the art;
- Feasibility study data science for Machine Learning algorithm, classification of images;
- Identification of the lack of available data to respond to the specificity of the business and especially the costs of repairs specific to the skills of employees and internal organization;
- Work on an intermediate solution to meet the expectations of improving internal organization while capitalizing on data recovery;
- Benchmarking of tools to label images, selection of the most suitable solution and development of optimizations to simplify the work of data labeling for the modeling stage in supervised learning;
- Preparation of Deep Learning algorithms based on images and models to be used, which will have to be refined on the final dataset, including a pre-processing phase. The modeling gives good results but must be adapted to the criticality levels of the client who will feed these dataset.