How We Taught Our Artificial Intelligence to Learn Retail Shelf Monitoring

Written by satyamkannaujia | Published 2018/09/20
Tech Story Tags: artificial-intelligence

TLDRvia the TL;DR App

Latest advancement in AI technology and deep learning algorithms are changing the retail industry. With large number of data sets comprising thousands of shelf images, companies can now leverage of AI to better monitor their retail shelf presence. AI will help in recognizing product conditions on shelf such as availability, assortments, space, pricing, promotions and many more. It will empower companies to take immediate corrective. AI algorithms can definitely improve planogram compliance by providing accurate stock visibility insights. Companies will be able to monitor and benchmark duration of out of stock instances, which will lead to better in-store product placement.

How will it work

Not much will change in the daily routine of the field agents apart from the fact that they will have more flexibility in terms of quality of pictures that they have to share with the analysis team. Current industry have a lot of bottlenecks that affects final insights in which failure to analyse unclear images is a major issue. This leads to increase in time and cost to the company to retrieve new images for fresh analysis. Field agents will just have to click pictures of all the relevant shelves and feed it to the AI system. Obstruction while the field agents click shelf pictures is another damper in the retail audit process. This too is taken care of by AI as the system becomes highly scalable and loss of pictures due to obstruction while photography can be ignored.

Source: Innovation is everywhere

The AI algorithm will analyse all types of inputs and deliver insights. Its capability to analyse poor quality images will enhance the credibility of the final results. Traditional systems have a hard time analyzing unclear/low light images which is not the case with AI. Confusion between similar looking products is another contentious issue which is resolved by using AI.

KARNA AI has leveraged the power of AI to create SHELF WATCH, an AI shelf analysis service that empowers field agents with flexibility and companies with scalability. Shelf Watch will eliminate all gridlocks in the traditional retail audit process that is currently eating into the revenue of the consumer goods organisations. The extent of its advantages can be fully understood by analyzing each stakeholder in the retail audit process.

Field Agents

The reps face major challenges while collecting data in the form of pictures and videos. There is lack of uniformity in stacking patterns across retailers which leads to different kinds of pictures in terms of stock orientation, lighting and positioning. Field agents struggle with maintaining consistency with the data they collect because such non-standard pictures take longer to analyse. In the pursuit of standard images, field agents fall prey to other types of human perception biases.

Shelf Watch helps the field agents by giving them the flexibility to take all possible pictures in any orientation, lighting or positioning. Such flexibility is allowed because shelf watch is not dependent on standard uniform images to give accurate output. Using state-of-the-art AI algorithms, Shelf Watch is able to analyse even the most distorted images because it uses AI pack recognition technology.

Retailers

Compliance audits are tough tasks for retailers as well. To comply with the pre-set planogram is part of the service agreement between the retailer and the brands. If in the final assessment the retailers are found to be violating the agreement by displaying too few products, or by not positioning the products correctly, can attract penalties and even termination of contracts ( in extreme cases ).

Since Shelf Watch allows field reps to be flexible while collecting data, it will also help retailers comply with the service agreements because all the images collected by the agents are analysed irrespective of the light, positioning and orientation of the products on the shelf. This saves retailers from false audit reports because even if their shelf is not well stacked in terms of positioning and lighting, Shelf Watch will detect all the objects on the shelf, thus reducing incidences of non compliance due to poor data collection.

Brands

Finally it will be the Consumer Goods companies that will benefit the most from our AI powered solution. They will be able to analyse all types of pictures from retail audits by using Smart Gaze for shelf object detection. Smart Gaze will help cut the time lag between input data and final insights. This abets the company to take on-time corrective action, if necessary.

If you want to know more about Karna AI and our industry solutions, please connect with us on our website.


Published by HackerNoon on 2018/09/20