Innovation Challenges

TVS

Operators perform tasks along with pre-programmed machines across assembly lines in a set of predefined sequences. Even though the current system has a well-established process, human errors still occur. In addition, the system lacks data, insights of real-time analytics of operator actions, and alerts for deviations in the process flow.

Visual inspection by operators varies from person to person, and thus could sometimes lead to a defective product.

Though computer vision has been in use in manufacturing for decades, recent advances in Machine Learning (ML) and image processing have enabled new use cases. No longer limited to structured, repetitive tasks. AI-assisted, high-level computer vision and Hand Motion Analysis solutions are capable of functioning in increasingly complex, high-speed manufacturing environments and in conjunction with operators.

The desired result is improved efficiency, fewer errors, and better data.

What We Are Looking For

TVS believes that industrial transformation is an opportunity to elevate the manufacturing facility using AI and ML-based computer vision and Hand Motion Analysis (HMA) to monitor the actions of an operator in real-time, and enable defect-free assembly lines to unlock the potential for long-term competitiveness and sustainability.

Nearly 63% of total deaths in India are due to non-communicable diseases, of which 27% are attributed to cardiovascular disease which affects 45% people in the 40-69 age group. High blood pressure is among the most important risk factors for cardiovascular diseases. Moreover, it remains poorly controlled due to low awareness about hypertension, lack of appropriate care through primary care and poor follow-up[3].

Hypertensive individuals adopting healthy lifestyles, such as being active for at least four days per week, weight-loss in the presence of obesity, consuming a diet rich in fruits and vegetables, and sodium below the recommended threshold, avoiding high alcohol consumption and refraining from smoking have been effective lifestyle changes to prevent or control early stage of hypertension[4].

What We Are Looking For

Apollo 247 is looking to decrease the number of patients with uncontrolled hypertension to improve each patient’s overall health and well-being. A common problem with such programmes is the lack of consistency in adhering to the proposed lifestyle changes. For example, patients do not take their medication regularly, have trouble sticking to their exercise regime, and deviate from recommended dietary changes. 

The desired solution could come in the form of hardware, software or a combination of both. There are many such health and wellness applications in the market, but a specific application focused on high blood pressure is preferred in the effort to better monitor, engage, and manage patients.

Motivation to follow the required workout routine and recommended diet, clock enough sleep, and take medication on time is generally lacking. The solution provider should take this into consideration and propose an effective solution to address this as well.

The solution should also cover the following requirements:

• Integration capability with leading activity and health trackers. Selection of trackers can be mutually agreed upon alignment of the pilot project. Apollo 247 will help with getting those trackers.

• A front-end to engage the hypertensives, and a back-end for care providers.

• Focus on the Indian and Southeast Asian market.

Skin diseases are more common than other diseases and have a serious impact on people’s life and health. Skin diseases may be caused by fungal infection, bacteria, allergy, viruses, etc. However, access to dermatology specialists and diagnostic processes is limited, time-consuming, expensive, and invasive even at early stages. While all countries may have challenges for access to treatment, emerging countries face a greater incidence of a variety of factors but much lower dermatologist-to-population ratios to treat disease.

Image recognition of skin disorders can help classify the disorder to help take appropriate action remotely, at an early stage, and in situations where there is lack of access to expertise. In the treatment of skin disease, early detection is the critical factor to cure the disease, effectively reducing its impact and improving quality of life, especially in patients with other underlying conditions such as diabetes and hypertension.

Separately, the accuracy of skin disease recognition is inconsistent due to the similarity between different skin diseases and the limited number of dermatologists with professional knowledge. Nowadays, the majority of conclusions on the patients’ existing symptoms are drawn mainly based on doctors’ years of experience or their own subjective judgments, which may lead to misjudgments and consequently delay the treatment of the condition. To solve this challenge, the algorithm should be highly accurate.

What We Are Looking For

Apollo 247 is looking to improve overall patient engagement in dermatology by providing easy, low-cost, quick, efficient, and high-accuracy diagnosis of skin conditions with the convenience of patients’ own smartphone cameras. This will help in delivering consistent quality healthcare for patients who do not have immediate access to primary healthcare centres.

The desired solution should come with a strong image processing capability powered by machine learning algorithms to effectively analyse the image of affected skin collected through a smartphone camera. It should classify the type of skin condition and report back to an appropriate dermatology specialist to plan the next steps.

The solution should also cover the following requirements:

• A micro-service API and a front-end interface;
• A multicast classification with high-precision and recall of conditions analysed;
• The MVP can be a mobile app with an ability to identify top 10 commonly seen skin disorders in outpatient settings with high precision and recall;
• Focus on the Indian and Southeast Asian markets.

In 2021 the Food & Agriculture Organization of the United Nations (FAO) estimated up to 40 per cent of global crop production is lost to pests each year. Plant diseases cost the global economy over US$220 billion, and invasive insects at least US $70 billion each year. Not only does this cause massive loss of income to farmers it is tragic when FAO estimates hunger affected between 700-829m people in 2021. As climate change is predicted to increase the impact of pests, the losses are likely to increase.

Decades of mono cropping and inappropriate application of fertilisers has impacted soil health, further exacerbated by water challenges. We need a better understanding of conditions to manage crop growth.

Many crops and plants are very sensitive to a range of internal and external factors that can cause them to perish very quickly. Soil parameters impact plant health directly and regular monitoring is paramount to alleviate possible problems.

Plant health management is currently based on human visual reporting and subsequent management. Human errors in timing and / or interpretation impact the timely detection of such issues. Experts in plant health issues cannot be present across all sites. Problems need to be centralised where identification, analysis, and resolution is possible with the best resources and capabilities. Further, past data and causal relationships can inform and support future actions , both pre-emptively and reactively to new incidents.

Advances in automated data collection and analysis, machine learning and the development of artificial intelligence and predictive solutions have improved efficiencies in operations across many industries. The same is possible and increasingly necessary in agriculture to prevent crop and plant losses to enable us to better feed the world.

DS Group is a major conglomerate with multiple locations across India growing crops including macadamia, blueberry, almonds and Stevia and currently expanding internationally.

What We Are Looking For

Utilise machine learning and AI with hardware to better understand soil and plant parameters to significantly reduce pre-harvest plant and crop losses caused by pests and disease.

What We Are Looking For

There have been several machine learning models that have tried to address the above problem or an allied problem of similar nature. However, most of the algorithms do not succeed in replicating the model across different data sets (standardisation of solution is not always achievable). HCG would like to take this opportunity to explore the most appropriate solution to address this problem and become a game-changer in this industry.

The desired solution should fulfil the following requirements:

• The solution combines feature extraction from medical imaging. It uses deep learning to fuse data analytics to iteratively optimise one with respect to the other. In other words, deep learning provides radiomics models with optimal features and optimal data analysis for a specific clinical problem.
• A deep learning algorithm that learns from existing datasets and suggests probable treatment protocol for similar patient characteristics.

The expected development roadmap is as follows:

• Phase I: Develop a response prediction model based on an existing dataset
• Phase II: Implement the data model on a real-life application

Capitaland has an extensive presence in India, operating multiple business parks, warehouses and data centres across the country.

They have several Safety Management Systems (SMS) in various facilities and development sites. These systems have been in place for the past five years to achieve zero fatalities and prevent serious injuries at their worksites.

However, as Capitaland expands its real estate portfolio, there has been a crucial need to build on these existing systems to shift from a reactionary approach to a more stable and scalable model that can be applied to all their pre-existing and upcoming facilities, through the adoption of innovation and automation.

The main cause for injuries and death in India is falls from height. Despite sufficient personal protective equipment and safety protocols in place, there are challenges in fostering a culture of safety amongst workers. Some of the current safety protocols include “Kaizen” initiatives, incentives for hazard identification, safety promotional activities, online work permits, and online training modules.

Facade cleaning is an area of interest. At present, operations are carried out using two methods:

1. “Spiderman”, which due to building designs and other external factors, is the primary method used by Capitaland for facade cleaning
2. Cradle or gondola

What We Are Looking For

Capitaland wants to develop a solution to improve safety for work-at-height, utilising existing technologies to protect workers and prevent fatal injuries. Proposed solutions could also utilise behavioural science to drive behavioural change and embed a safety-first mindset.

The solution should consider the following features:

• The reduction or elimination of the need of a human operator in current protocols;
• The application of computer vision technology, video analytics, and drone-based monitoring;
• Video analytics to track and understand operator behaviour; and/or
• The application of behavioural analytics to educate and train operators.
• For a facade-cleaning solution, consider that Capitaland’s facilities have a ratio of 60% flat and simple surface structures to 40% complex surface structures. Being able to cater to complex surface structures is a big plus.

Solutions that mitigate the risk of injuries/death from the above methods are welcomed.