Skip to content

Master project

The WattWise Project: Exploring Energy Transition Design

What Started It

The WattWise project, my Master’s thesis design initiative, began with a module in the first term of my program called “Living with Your Own Ideas.” This module was centered around a unique approach called first-person perspective research, which fascinated me right from the start. The concept was simple yet profound: to understand a problem deeply by experiencing it firsthand. This was a significant departure from my previous academic experiences, where issues were often explored from an outsider’s perspective, making this method both exciting and challenging.

What Was the Original Interest

My interest was initially piqued by a book titled “The Ministry for the Future” by Kim Stanley Robinson. One of the characters in the book attempts to live by the principles of the 2000 Watt Society, a real-world concept aimed at reducing individual energy consumption to 2000 watts per day to promote environmental sustainability. This idea intrigued me because it proposed a way to live sustainably without drastically affecting quality of life. I was eager to explore if this balance could be achieved and how it might be facilitated through design.

What Was the Problem We Were Looking Into

The core problem we wanted to tackle was overconsumption of energy in modern societies. The 2000 Watt Society’s goal is ambitious, aiming to reduce energy use in a way that doesn’t negatively impact day-to-day living. The challenge was to find practical methods to help people reduce their energy consumption while maintaining their standard of living. This involved looking into how energy is consumed in various aspects of life—home, transportation, and food—and identifying ways to make sustainable living easier and more appealing.

Partnering with Carlotta

As I dove deeper into the project, I connected with Carlotta Hilkema, a classmate who shared my enthusiasm for sustainability. Carlotta was particularly interested in behavior change and how design could influence sustainable choices. Our shared passion for these issues led us to collaborate on this project. We believed that combining our strengths and perspectives would allow us to tackle the problem more effectively.

What Were the Different Angles We Approached the Problem From

We approached the problem from multiple angles:

  1. First-Person Perspective Research: We began by living the principles ourselves. I undertook a 24-hour challenge to track and reduce my energy consumption, meticulously logging every watt used in my daily activities.

  2. Technology and Tools: We explored existing tools and developed new ones to help track and manage energy consumption. This included creating a Google Sheets tracker and building a hardware energy monitor paired with an IoT system for real-time feedback.

  3. Big Data and AI: We integrated big data tools and AI to automate data processing and provide actionable insights. This led to the development of WattWise AI, which used a conversational interface via Telegram to make energy data more accessible and engaging.

What Did We Discover

Our discoveries were multifaceted:

  • Tracking Energy Consumption: The initial challenge revealed that manually tracking energy consumption was incredibly time-consuming and complex. It was clear that any practical solution needed to simplify this process significantly.

  • Behavioral Insights: We learned that sustainable behavior change requires making sustainable choices as easy and convenient as possible. People are more likely to adopt new habits if they don’t feel like a burden.

  • Technological Potential: The integration of AI and big data showed promise in automating complex calculations and making sustainability data more user-friendly. However, it also highlighted the need for user-friendly interfaces and accurate data sources.

  • User Preferences: Simplicity and relatability in tracking methods were crucial. Users preferred straightforward comparisons, such as energy usage in terms of smartphone charges.

  • Privacy Concerns: Ensuring user comfort with data privacy was essential to gaining trust and engagement.

  • Engagement: Human-like interactions were more engaging and effective in encouraging habitual change, highlighting the importance of a user-friendly interface.

What Worked

Several aspects of our approach proved successful:

  • Google Sheets Tracker: Initially, the tracker helped in understanding energy consumption patterns and was useful for a few days.

  • Energy Monitor and IoT System: The hardware we developed provided real-time data, making it easier to see the immediate impact of energy use and reduced the burden of manual input.

  • AI Integration: Using AI to process data and provide feedback through a conversational interface made the information more accessible for users. This bot we developed provided energy usage tips based on user inputs like photos and voice memos, making the process engaging and easy.

What Didn’t Work So Well

We encountered several challenges and limitations:

  • Manual Tracking: The Google Sheets tracker became tedious after a few days, and reliance on extensive user input proved impractical and cumbersome, leading to user fatigue and decreased engagement. This was an important learning for us as it really showcased the need for a more automated solution.

  • Overwhelming Data Collection: Collecting and logging data manually was impractical for long-term use and detailed tracking methods were overwhelming for users. We needed a more seamless and less intrusive method.

  • Complexity of AI Systems: While AI showed potential, developing and refining the algorithms required more resources and expertise than anticipated.

Did We Pivot, If So Why

Yes, we did pivot. Initially, we focused on a detailed tracking system for energy consumption at home. However, we realized that this approach was too narrow and didn’t address the broader picture of sustainable living. Our pivot involved expanding our scope to include a holistic view of energy consumption across home, transportation, and food. We aimed to create an integrated system that could provide comprehensive insights and recommendations, making sustainable choices easier and more practical for users.

What Did We End Up With

We ended up with the WattWise project, a holistic tool designed to track and visualize embodied energy consumption and carbon footprint across various aspects of daily life. The core of the system is an AI-powered platform using the GPT-4o architecture, accessible through a Telegram chat interface. This system allows users to log their energy use, receive feedback, and get personalized recommendations for reducing their consumption.

We also developed supporting hardware, including an energy monitor, to facilitate data collection. This energy monitor sends its data to the AI assistant, enabling it to provide accurate and tailored advice based on real-time information. Additionally, we experimented with other tools like a transportation tracker to give comprehensive insights into various aspects of energy usage.

This approach combines automation with user engagement, addressing both the knowledge gap and the practical challenges of energy tracking. The Telegram bot offers a human-like interaction that is easy to integrate into daily life, while the energy monitor simplifies data collection. This dual approach ensures that users can easily understand and manage their energy consumption without feeling overwhelmed.

Why Did We Choose This

We chose this direction because it aligns with our goal of making sustainable living practical and accessible. By leveraging AI and big data, we can automate complex tasks and provide users with clear, actionable insights. The conversational interface makes the system user-friendly and engaging, encouraging continuous use and fostering sustainable habits. Our focus on open-source solutions ensures that the project remains accessible and adaptable, allowing for community involvement and further development.

Where Do We Want It to Go Next?

Broader Testing: We aim to gather more feedback from a wider audience to refine the tool and ensure its effectiveness across diverse user groups.

Gamification: Exploring gamification strategies to increase long-term engagement and make energy tracking more enjoyable.

Business Applications: Considering a shift towards lifecycle assessment tools for designers, providing a valuable resource for sustainable product development.

Scalability: Transitioning to open source models like Llama-3 or Phi-3 to handle the inputs and outputs to make WattWise more accessible and adaptable.

Looking ahead, we aim to refine the WattWise system, addressing any bugs and enhancing its features. We plan to expand the AI’s capabilities and integrate more data sources to provide even more accurate and personalized recommendations. We also want to explore running the system on local machines using open-source AI models to reduce its environmental footprint and increase privacy for users. Additionally, we hope to build a community around the project, encouraging collaboration and sharing of sustainable practices. Ultimately, our goal is to create a tool that not only helps individuals but also supports businesses and larger communities in their transition to more sustainable practices.