Community Voices: How Constems-AI Systems addressed the problem of foreign particle contamination in processed foods

Amit Singh and Amit Srivatsava launched Constems-AI Systems in 2017 to usher in innovation in sectors that had increased human-machine interactions. They wanted to do that by introducing intelligent automation based on computer vision and image analytics to improve the consumer goods product quality, availability and distribution across the business process value-chain.

The duo had noticed that consumer goods companies face challenges in tracking, tracing and maintaining quality of products across the process value chain. Keeping a tab of these is difficult as companies have 1000s of stock keeping units (SKUs) in their brand portfolios. Most of these companies are dependent on human-based information to track and trace product availability or placement on-shelf.

Constems-AI Systems has been countering these challenges by their efforts in the computer vision and image analytics space with their AI-enabled SaaS platform CAInatics. It offers state-of-the-art metadata based image analytics and AI-enabled recommender SaaS application, which can be accessed via software as services by the enterprises. CAInatics can be easily integrated, making processes and decision making faster, intelligent and highly accurate.

Making a difference

Through their technology, Constems-AI Systems has been targeting two distinct sectors - retail and manufacturing.

In retail, the company aims to help across processes like distribution, availability, complaint and sales ROI. Its goal is to digitize brick and mortar retail by image analytics SaaS platform enabled with multi-model and Advanced Analytics for better space planning, share in Retail, retail asset management, compliance audit and retail execution. This will help enhance sales and drive marketing efficiencies.

For manufacturing companies, the team’s target is quality related KPIs. In the sector, dependency on human quality checks can lead to massive implications in terms of cost and bad customer experience. CAInatics, their easily deployable system with Edge compute, helps automate QC processes with end-to-end tracking of production output to inspect and eliminate product defects possibilities. The solution also has real-time OEE dashboards and RCA (Root Cause Analyses) Tools for quality teams to control production quality related issues.

Since their inception, the team has had to overcome challenges like companies’ openness to adopt new-age technologies. Despite this, they managed to build a long-term relationship with customers like ITC, P&G, J&J, Reckitt, GSK, PepsiCo, Mahindra, Adity Birla to name a few.

The Constems-AI Systems edge

Looking back, the team reminisced about an interesting case study in the beginning of the organisation’s journey. A leading global agri product player was struggling with foreign particle contamination in their end products.

Most food and agri-based manufacturing plants try to enforce robust quality checks on their products, and they utilize a trained team for that. But given the scale of production at these plants, product QC inspections are not only limited to a small sample of production, but also highly dependent on manual knowledge and subjectivity. Additionally, these products are expected to meet intricate quality standards and to assess these QC standards with consistent results and rejection in the real-time is a bigger challenge in existing processes at plants for the agri-product industry.

To counter this, Constems-AI Systems proposed an AI-based quality grading solution. They conceptualized, designed and deployed Computer Vision-based AI Systems to conduct quality tests (physical parameters) for agri-based products on the running production lines and conveyor belts. The solution was deployed in conjunction with an advanced Robotics Process Automation (RPA), which was a rejection system. The standard three parts to deploy such an intelligent solution on the factory shop-floor are vision-based industrial camera, on-premise edge system and real-time reports.

The AI-based visual inspection system was designed to address the problem of foreign particle contamination in the processed foods and agri-products on the running (manufacturing) line for the customer. Constems AI Visual Inspection platform used its monolith architecture to provide AI on Edge with Cloud-based architecture, which is easily deployable at fast moving manufacturing lines. This system was used to address the given problem statement. Through this, the team could inspect the quality of more than 20 types of foreign particles per 100 milliseconds. Additionally, it helped the customer save more than 70 percent of the cost of quality inspection and improve the product quality reaching the end consumers.

The way ahead

As Constems-AI Systems continues to chart an upward growth trajectory, they plan to expand deeper in South Asian and US markets in the coming years. Constems AI already working on deploying product solution beyond the visual spectrum like Thermal imaging, Hyperspectral, and X-rays.

About Community Voices

Community Voices is a series of articles from the Plugin Alliance that spotlights solutions and startups helping India’s manufacturing and supply chain industries benefit from smart technology. The series leads into Innovent 4.0, an exclusive industry technology conclave from the Plugin Alliance — a first-of-its-kind Industry 4.0 alliance in India — with the aim to showcase India’s role in the Fourth Industrial Revolution. In tune with the event, the series will continue to highlight stories from Industry 4.0 startups.