Exclusive Interview with Sunil Kardam, Supply Chain and Logistics SBU Head and Client Partner, Gramener

The data science industry suffers from one certain anathema, i.e., information overload, which not all data science teams can address without losing the bigger picture. Gramener has identified this gap in the data science project management and precisely has been responsible for the company's inception as a Data visualization company. Later foraying into analytics Gramener has made a mark for itself in helping companies with domain-specific analytics. Analytics Insight has engaged in an exclusive interview with Sunil Kardam, Supply Chain and Logistics SBU Head and Client Partner, Gramener.

1. Kindly share your point of view on the current scenario he Big Data Analytics and its future.

Today, there isn't any industry that is not influenced by Big Data & Analytics. The realm of possibilities is innumerable. The manufacturing sector, that once considered laggard in embracing data culture, is increasingly being transformed by Industry 4.0 revolution – at the center of which sits data analytics. Digital Twins are the driving force to link key parameters across the equipment and optimize the production line. Using it, the equipment failures can be predicted ahead of time. Computer vision made it possible for the Pharma industry to identify defective drugs or faulty packaging on a near real-time basis and releases alerts to the line managers for suitable interventions. Geospatial analytics has wide applications for Sustainability vertical and has become one of the favorite investment portfolios across the board. A lot of it has to do with compliance & regulation. This technology has shown positive outcomes to identify pockets of heat islands or recognize patterns for the population spread for a country or a state, thereby allowing governments to institute the necessary interventions. A case in point is Nisqually Foundation collaborating with us to deploy a computer vision solution for species detection. It is evident that structured data is no more the sole battlefield in the market anymore, and enterprises are increasingly looking at unstructured analytics techniques such as NLP & NLG to stay relevant and march ahead of the competition.

2. Kindly mention some of the major challenges the company has faced till now.

Data analytics despite being a vibrant field, we faced numerous challenges to come up with a repeatable & scalable solution. First & foremost, the industry definition of analytics is quite wide, and it varies from simple reports/dashboards to complex predictive analytics. Client maturity is still picking up. We started as a visualization company and gradually moved gears to core analytics. With limited client exposure, it became difficult to communicate our change of focus to clients and numerous old customers still come to us asking for visualization services.

Another challenge is to identify domain solutions with a high degree of repeatability. Customer mostly looks at the solution as a product and prefer the SAS model that remains light on the pockets too. Building a solution with good repeatability need significant capital and a great deal of time investment to achieve perfection. It is easier said than done. The proliferation of analytics-driven SAS solutions are yet to see the light of the day.

For our solution to remain highly relevant in the market, it needs good collaboration with business. It is not always easy to get this part right considering data scarcity, inadequate business knowledge, even on the client side, and a lot of unknowns surrounding it. We have come a long way and bridged most of this gap but occasionally we do get surprises.

3. What is your biggest USP that differentiates the company from competitors?

We clearly prioritize business over technology, and not the other way round. Most enterprises do not get this part right. Simply pumping money into analytics with a weak link to business will not yield the desired result. We have invested heavily in domain expertise to bridge this gap.

Our second differentiation lies in Gramex. It is a low code python-based data platform that comprises 200+ microservices. It alleviates the need to build an application from the scratch thereby reducing development time by up to 40%. The costs saved are indeed gain for customers.

We just don't deliver analytics solutions but implement an adoption plan around them. With most enterprises still catching up with analytics, an adoption plan is pivotal to building the right data culture and data strategy for any enterprise.

Further, all our solutions are supported by a robust business case, and we enable clients to track the business benefits as projects are rolled out. It improves transparency and helps clients seek subsequent funding to roll out high-impact use cases.

4. Tell us how your company is contributing to the IoT/AI/Big Data Analytics/Robotics/Self-Driving Vehicles/Cloud Computing industry of the nation and how the company is benefiting the clients.

At Gramener, we have a very specific focus on technologies that can bring transformational change. This includes digital twins, computer vision, geospatial analytics, NLP/NLG, etc.

For example, digital twin solutions have a great potential to transform the pharmaceutical industry by mimicking the process line. Similarly, manual inspection takes place in almost every industry. Computer vision can increase its accuracy up to 99.99% by automating the process using image processing.

About 40% of the global population is vulnerable to deadly mosquito-borne diseases like malaria, chikungunya, yellow fever, zika & dengue. The World Mosquito Program (WMP) uses natural bacteria to modify mosquitoes, reducing their ability to carry deadly pathogens. Their off-springs also lose the ability to spread diseases. It is vital that these mosquitoes be released in areas where they can reproduce rapidly.

Under the Microsoft AI for Good grant, Gramener partnered with World Mosquito Program (WMP) to provide a solution. Using computer vision models on high-resolution satellite images, we identified population densities that were at risk from mosquito-borne diseases at a sub-neighborhood level. The WMP team used our AI-driven release-and-monitor plan, cutting down the time it takes to identify release points from 3 weeks to just 2 hours & increasing accuracy by up to 70%.

The WMP solution also employed geospatial AI, using satellite imagery to analyze population density & identify the best neighborhoods to release the mosquitoes.

Gramener also used geospatial analysis to identify Urban Heat Islands (UHI), areas with temperatures higher than their surroundings, in Calgary, Canada. Identifying these hotspots poses a challenge for local authorities as they require a lot of data & analysis.

Under the Microsoft AI for Earth grant, we collaborated with Evergreen, a non-profit, to develop a solution that used Big Data, AI & Geospatial Analytics to enable municipalities to plan effective climate interventions & take remedial measures at the street level, helping to create a resilient city.

In 2021, on the eve of cyclone Yaas in India, SEEDS, the disaster management non-profit, deployed a geospatial solution developed in conjunction with Microsoft & Gramener to identify risk-prone households directly in the path of the storm. The application achieved a staggering 90% accuracy, helping SEEDS issue prior warnings & successfully evacuate 1,100+ families.

5. How is IoT/Big Data/AI/Robotics evolving today in the industry as a whole? What are the most important trends that you see emerging across the globe?

One of the most noticeable trends is increased focus on data analytics. Having said that, many companies have not yet jumped on the bandwagon. They don't know where to start.

Thanks to publications like Forrester & Gartner, there is a lot of awareness around data analytics. But one of the biggest hurdles that companies face when implementing data analytics is the lack of data. Even if the data is available, its quality is suspect.

Most industries are aware of the benefits of data analytics. However, they face challenges in implementation. This includes the pharmaceutical industry, FMCG, retail, manufacturing, etc.

Most of these sectors possess humungous amounts of data. Successfully using this data to generate actionable insights that lead to long-term growth is where data analytics comes in.

Previously, companies were heavily invested in reporting – KPIs, dashboards, etc. Today, they are on the lookout for tools that will help them diagnose, predict & prescribe, using the same reports. They want to use the data to generate actionable insights.

Earlier, businesses dealt with large amounts of data that was unstructured. Today, they want to go beyond insights & make the data actionable.

For example, if a company owns 5 plants & one of them is not delivering from a cost standpoint, data analytics can provide answers that can help bring down the costs at the troubled unit.

One important trend that emerges from this is the need for actionable reporting. In addition to analytics & algorithms, companies are also relying on hardware such as IoT to capture as much data as possible.

Before, large multinationals were expected to invest in technologies like analytics. Today, smaller organizations have also realized the advantages of applying data analytics to their processes.

For example, applying computer vision can identify product defects, improve overall quality, reduce costs & make manufacturing more efficient. Even startups with limited resources understand the long-term value of data analytics & its potential contribution to their growth stories.

Global corporations are also embracing data analytics, increasing their focus on data culture & making their organization more data-driven. Data analytics is not just the simple application of analytics to data sets. It also involves a change in mindset & skill upgradation.

This requires a change-management drive at an organizational level, necessitating heavy investments.

6. What are the key trends driving the growth in Big Data analytics/AI/Machine Learning?

Currently, there is an increased focus on making decisions based on data rather than intuition. With the world steadily moving towards predictive & prescriptive solutions, the one trend that is slowly emerging out of this mix is the democratization of data.

In an ecosystem where data is democratized, decision-making power is not restricted to one or a few people. Individuals in the middle & lower management can also make calls in their day-to-day operations.

To successfully implement analytics, there should be well-integrated systems in place, especially in the supply chain industry. For analytics to work effectively, the supplier, manufacturer, distributor & retailer have to be integrated & connected.

Without holistic integration, analytics fails to give us the big picture which, in turn, does not lead to big dividends. It may also lead to the problem of sub-optimization, where solving one problem on a small scale creates more problems elsewhere.

Another important emerging trend is the adoption of industrial IoT or IIoT. Earlier, when faced with a hardware issue, businesses would either replace the device or run a Six Sigma project. Today, they can add a sensor to the device to collect critical data. They can then run sophisticated algorithms using this data.

7. How is your company helping customers deliver relevant business outcomes through the adoption of the company's technology innovations?

At Gramener, we focus on front-end technology areas, providing predictive & prescriptive analytic solutions like digital twin, computer vision, NLP, NLG, geospatial analytics, etc.

GRAMEX, our low code platform, comprises 200 microservices or modular components that can expedite process delivery by up to 70%, helping spur innovation with faster go-to-market strategies.

We are also focused on domain expertise, connecting business with analytics. Our motto is business first, analytics second. To provide an effective solution, we first try to understand the business model of our client – a key aspect many analytics companies miss out on.

We try to link every decision we make & every analytics implementation idea we propose to business advantages. Our objective is to deliver RoI & create a plan around the business benefits our solutions offer.

Today, many companies are investing in analytics. However, they often don't have clarity on the advantages that result from these solutions. Helping a client accurately gauge the business benefits of implementing data analytics solutions is sometimes a project itself.

8. How C-suite executives can leverage data to deliver business value to their organizations?

To implement effective data-driven solutions, business leaders must first evaluate their organization. They have to assess the amount of data their operations & processes generate, and the quality of the data.

Once the quality of the data collected has been assured, the company can identify the data pipelines or repositories where all the data can be consolidated. This process usually takes weeks or months.

Once companies have established the 3 prerequisites to implementing a viable data analytics strategy – quantum of data, quality of data & data repository, they can leverage applications such as GRAMEX to mine more value from the data using analytics.

Organizations can build data mining capabilities in-house, often referred to as Data Science Centers of Excellence (CoE). They can also partner with data science leaders such as Gramener to run their analytics initiatives.

To ensure the successful adoption of data analytics solutions across an organization, C-suite executives have to create buzz & excitement around the idea from the very beginning. One way of generating engagement & interest among employees is to let them come up with innovative names & terminologies for the project.

Individuals whose work will be affected by the implementation of the data analytics solutions should be allowed to participate in every stage of the implementation process. This will help build their knowledge & training around analytics, improving their data literacy.