How ML and IoT can Revolutionize Waste Management in India

By: Tisha Sharma

We are no stranger to the fact that Covid-19 has had a grave impact on our society. Its effect can be seen across mental health, education, economy, politics, and waste management is no exception. But why waste time on waste when we can talk about cryptocurrency or Windows 11? Firstly, there are numbers to prove that the waste problem is alarming and secondly, I have nothing better to do.

According to the data collected between June and December 2020, 32,996 metric tonnes of biomedical waste was produced in India. 70% of it was Covid-19 related waste which included used PPE kits, masks, shoe covers, gloves, swabs, etc. In another study, the International Institute of Population Sciences (IIPS) has found that the risk of infection has increased in 23 states due to improper disposal of bio-medical waste. Ultimately, it is the sanitation workers who have to skim through heaps of garbage with inadequate protective gear, thus risking their health for a living.

Now, what is waste management? Waste management involves handling waste from its source to its final disposal. This includes the collection, transport, treatment, and disposal of waste while adhering to waste-related laws. I first started reading about this for a hackathon where we had to come up with a solution for medical waste disposal. My team realized that most municipalities were falling short in managing waste, due to a lack of equipment, infrastructure, and proper planning. To tackle this we came up with a prototype where waste disposers and collectors could register and connect through a website. One of our prospects was to make use of image detection to segregate waste into different bins.

A similar idea has been implemented by an Ahmedabad-based startup, Ishitva Robotic Systems (IRS) with their IoT-enabled device, ‘Sanjivani’, which segregates recyclable waste. Sanjivani consists of 3 products– NETRA, YUTA, and SUKA. NETRA is a computer with industrial cameras, sensors, and neural networks that learn about waste by taking images and detecting recyclables on the conveyor belts. YUTA, an AI-powered robotic sorter, uses NETRA’s vision system to lift and segregate the recyclables on a high-speed conveyor belt. Lastly, SUKA, a modular sorting machine, makes use of AI-powered air sorting techniques. While most medical waste isn’t recyclable, we can use image detection techniques similar to Sanjivani to separate hazardous waste which can be chemical, radioactive, infectious, etc, from non-hazardous waste like packaging or food wastes to ensure safe disposal to landfills or incinerators.

Another notable example is Songdo, in South Korea where electronic food waste bins are equipped with RFID (Radio-Frequency Identification). Citizens have to tap their identification cards before they throw their waste into the public garbage bin. The garbage bin will then weigh the trash, calculate and collect it, and then send the garbage bill to the disposer. It uses a reader built into the automated pneumatic garbage disposal system without the need for manual collection or sorting. The waste is processed separately, buried, recycled, or used as fuel based on the data encoded in the tag. This is an excellent way to get the general public to participate in environment-friendly practices and contribute to a more sustainable future.

The last example I want to talk about is BioHitech’s Chatbot called BioHitech Alto, which uses Machine Learning to predict food waste trends. It can analyze a wide array of machine data and helps in understanding machine status via push notifications giving real-time information. This cloud-based platform can be utilized by businesses to get precise metrics on the type and volume of wastes, and also avoid unnecessary food waste by placing informed supply orders.

The primary purpose of using IoT-based Machine learning in waste management should be to save labor, the total cost of transport and transfer, and improve the general quality of life. Computer systems can perform functions like classification, pattern recognition, etc. by training them based on data collected by IoT devices and then make decisions based on new data. We can already see that AI, ML, IoT, and automation are being implemented in India’s major industries. I believe that automating waste handling will ensure proper segregation and disposal while reducing health risks posed to garbage collectors and sanitation workers.

But despite all of this, I still think the best way to tackle waste handling is to minimize waste in the first place, which needs to be dealt with by industries head-on. We, as consumers, can do our part by buying products that minimize the use of non-biodegradable packaging and reuse and recycle whatever we can. If we keep ignoring the current situation, who knows the movie WALL-E might become our reality. We would have to abandon Earth, contaminated by garbage, and move to space, which isn’t the most cost-effective option.

The amount of plastic waste generated in the country is estimated to reach 200 million metric tons in 2041. It is quite evident that existing waste management laws which are ineffective even now will be of little help twenty years into the future. IoT opens up a world of new possibilities that can give rise to more sustainable solutions and combat the rising environmental concerns in our country.

References:

Related posts:

We spend huge chunks of our lives on the computer or our cell phones. Writer’s especially are vulnerable to cybersickness, a condition that arises from too much time staring at screens and scrolling…

With Opening Day for MLB’s 2019 season last week and seeing all the fans out in their jerseys, it reminded me of how devoted the Blue Jays fans can be. The devotion and experience of baseball acts as…

Ivory towers appear so unbreakable —  A Tanka Poem