AI-Eco Sort Station

Let AI do the Sorting, you just throw the trash!

Why AI

I believe the AI should "Streamline and enhance efficiency in tasks that are either undesirable or require precision beyond common human capabilities."

Automating Unpleasant Tasks:
Using AI to handle waste sorting, a traditionally unpleasant and tedious task.

Enhancing Sorting Accuracy:

AI-driven image recognition technology ensures high accuracy in identifying and categorizing waste materials.

Data Collection and Optimization:

Collects valuable data to optimize waste management strategies, collection routes, and schedules

Background

As environmental awareness grows, smart waste sorting has become a crucial component of urban management and sustainable living...

In high-density areas (such as Midtown Manhattan), although the number of trash bins is relatively high, it may still be insufficient to meet the demand from the large volume of foot traffic, potentially leading to overflowing bins. Trash bins in high-density areas may require more frequent cleaning, while those in low-density areas might be overlooked in waste collection schedules.

How the Garbage and Recycling System Works in the United States?

Insight

Due to factories being located in different areas, waste must be transported through transfer stations and MRFs before it can be processed into raw materials.

There is a shortage of opportunities to educate the public. A major challenge in waste management stems from the incorrect sorting of waste, which complicates the recycling process.

Places that require sorting are in great need of manpower while also demanding efficiency. (Even AI sorting is applied)

Brief

AI-powered 'Smart Bin' automatically sorts waste.

Opportunity

Cost Savings:

Lowers waste management costs by reducing manual sorting.

Reduce Disposal Costs at MRFS:

Pre-sorting waste reduces the need for further processing at recycling centres.

Education opportunity

To educate users about proper waste management, recycling practices, and environmental sustainability.

DEMO

Insights

1. Training Set Size for Demo

While the training set is limited, it is sufficient to demonstrate the feasibility of the trash classification method.

2. Overfitting with Promising Generalization

Some data categories show signs of overfitting; however, the system performs well in identifying new items, showcasing its adaptability.

3. Proven High Accuracy

Research indicates that the trash classification system achieves an accuracy rate of over 90%, validating its effectiveness and practicality.

Goal

A form that is instantly recognisable, approachable, and clearly distinct from existing designs.

Shape

Make it easy to identify as a trash bin to avoid confusion.
Create a friendly, inviting look that encourages use.
Add a distinct silhouette or key feature so it stands out.

Mechanism

Use simple innovations that improve function (e.g., sensor lid, sorting compartments).
Keep interactions intuitive with minimal learning.
Balance innovation with practicality: efficient, easy to use, and easy to maintain.

Storyboard

Promoting Environmental Behavior

Making Waste Disposal More Engaging

Encouraging Community Participation

Secarnio

The colors of lights change according to trash categories

NightLight

When Metal is recycled

When Plastic is recycled

Optimizing Space Utilization:
The design maximizes available street space, making it ideal for crowded urban environments by keeping streets cleaner and less cluttered.

Public Health Improvement:

Improves public health by reducing risks of contamination and pests associated with poor waste management.

Positive Environmental Impact:

By increasing recycling rates and reducing contamination, the bin reduces landfill use, lowers pollution, and conserves natural resources.