Automation Can Mitigate Risks in Public Feeding

In a world that is increasingly adopting efficiency as the core mantra, Artificial Intelligence (AI) and automation have become the brainstem driving most activities, from education to corporate actions and public safety. So can we consider that the time is right to add AI to improve the efficacy of our public nutrition programmes?

Historically, there has been no set definition for AI because emergent technologies imply that the life of AI is similar to that of changing seasons. But when AI is looked at through the prism of singular use, like AI for food safety, we can begin to assign it a defined role: using machine learning to detect and predict risks based on data sets acquired through digital surveillance of the factory floor. Automation, machine learning, AI—all these systems adopt a strategic and straightforward philosophy of using digital coverage to create data sets that can improve operational efficacy. This mechanism enables the system to do what Larry Brilliant, Chief Philanthropy Evangelist for Google.org, the charitable arm of the technology giant, wished for: “early detection, early response”[1].

Nevertheless, despite the overwhelming evidence pointing towards the need to automate more and more surveillance functions, there has been an equal amount of doubt concerning the accuracy and transformative power of such systems when they are tasked with surveillance of food safety practices. The cost of implementing such strategies has also been a topic of debate, particularly in nations that are still developing or are underdeveloped and where the need for such mechanisms in public feeding is exceedingly essential.

But, as the pros are weighed against the cons, evidence strongly suggests that automation has a consequential role to play in modernising and improving the efficacy of existing food safety nets, especially in a post-COVID world where food contamination risks must be eliminated.

The past year has been a wake-up call for humanity as the pandemic ravaged multiple sectors, bringing us face-to-face with pressing social inequities that require immediate redressal from rising unemployment levels to gaps in our healthcare system to a rapid increase in poverty levels. The most daunting issue is the disruptions to the nutritional security of those who benefit from the nation’s large-scale food security programmes, like the Mid-Day Meal (MDM) Scheme, ICDS initiatives and Anganwadi programmes.

The MDM Scheme, the world’s largest school feeding programme, is tasked with upholding the Right to Food and Right to Education of the about 115 million children in government and aided schools. For many of them, these school lunches are the one nutritious meal that they receive every day. The Scheme has been extremely efficient in tackling the magnitude of the child nutrition crisis by addressing classroom hunger, particularly in children from socio-economically challenged families. Since the Scheme was launched, the percentage of malnourished children has been steadily declining, falling to 14% as of 2019[2].  

But the twin pillars that support the development and growth of future generations—nutrition and education—have suffered largescale disruption over the past year due to nationwide school closures. Uncertainty continues to cast a shadow on the MDM Programme owing to the staggered reopening of schools. As of March 2021, the World Food Programme (WFP) estimates that over 90 lakh[3] children are missing out on school meals in India (of which 52% are girls). But the Central Government’s decision to ensure that at least dry rations are provided to MDM beneficiaries in the interim has helped steady the shaking grounds beneath the feet of millions of children.

As individual states in India reopen the school gates, it is necessary that public feeding initiatives, such as the MDM Scheme, restart with enhanced food safety mechanisms, particularly in supply chain management and food preparation. In facilities where mid-day meals are prepared, like centralised kitchens, there is a need to reinforce personal hygiene measures in food manufacturing spaces and personnel because of the extremely virulent nature and mutability of the COVID-19 strain. 

Even before the pandemic, kitchens that were tasked with the preparation of mid-day meals successfully developed hygiene protocols for safe kitchen operations with the intent of eliminating the risk of food contamination. This included sanitation practices, worker hygiene requirements, disinfecting spaces and usage of personal protective equipment (PPE). However, with increasing pressure on enacting stringent safety measures in a post-COVID world, it has become the need of the hour to employ efficient mechanisms to monitor SOP compliance in public feeding initiatives.

The issue of contamination at any level of the supply-chain poses a considerable risk in a post-COVID world, where any incident can culminate in shutting down the entire factory line. Early monitoring mechanisms that rely on human effort have come under the radar of safety and hygiene compliance committees that posit that this system has a two-fold impact. Along with the expense of human supervision in large-scale settings, the human brain is susceptible to fatigue. So, a supervisor checking the surveillance video footage will miss most of the activity after just 22 minutes into viewing it. In a large-scale feeding setting, such an error will culminate in long-term disruptions to the nutritional needs of millions of children. 

An optimal mitigation strategy should be resilient and involve identifying all vulnerability risks while addressing those with a higher probability of recurrence. Studies show that implementing innovative, CAPEX-friendly technology could be the solution to ensuring efficient hygiene and safety compliance monitoring for large-scale public feeding programmes in a post-COVID world. Technology-driven interventions have the potential to reduce human error and handle unforeseen disruptions by automating space-people-processes and tracking supply chain risks in indoor environments.

One study has proven that using cloud-based solutions to detect ‘objects of interest’ can be an effective monitoring solution. This system would rely on AI-based video analytics that would leverage existing structures, like CCTV and Network Video Recorder (NVR) feeds, from the factory floor to detect variations in the use of protective PPE, social distancing and disinfecting of spaces. This would also make contact tracing much less cumbersome, thereby enhancing operational efficiency. The resilience and sustainability of using AI-enabled CCTV to track indoor environments are reflected in its promise of error-free detection of non-compliance in real-time. This will aid in rapid recovery by localising the impact instead of having to institute a full factory shutdown.

In a post-COVID world, the need for robust monitoring mechanisms in public feeding domains has become necessary. By addressing human limitations through preventive and reactive automation strategies in large-scale public feeding endeavours, it is possible to improve the delivery of essential public services. The skilful addition of automation to the mix can help the entire nation better address the gaps that might have arisen owing to the COVID-19 pandemic, taking us one step closer to releasing the UN Sustainable Development Goal (SDG) 2 of Zero Hunger.

(The author is the CEO of The Akshaya Patra Foundation. Views expressed are his own, and do not necessarily reflect that of Outlook Magazine)

 [1] Kristen M. Altenburger, K. M. A., & Daniel E. Ho, D. E. H. (2019, December 16). Artificial Intelligence and Food Safety: Hype vs. Reality. Food Safety Magazine. https://www.food-safety.com/articles/6416-artificial-intelligence-and-food-safety-hype-vs-reality