in|situ| Lab - Log In

Supporting decision-making for effective residential energy consumption

Internship proposal at in|situ|

Principal advisor: Theophanis Tsandilas (
External collaborator: Georgios Chalkiadakis (

Energy systems will soon converge to smarter and greener energy management and tariff solutions. The “smart grid” represents the vision of future energy systems where new energy sources, e.g., a small solar or wind plant, participate at any moment to satisfy an increased demand while detailed information about offer, demand and pricing flows in real time and in any direction between producers, energy providers and consumers [1]. Smart meters that allow energy users to easily monitor and understand their consumption patterns have already started being installed in several European countries, including France [2].

In this new reality, households will have to radically change their attitudes towards energy consumption. Producers and consumers will have to coordinate with each other to optimize energy management by minimizing costs and prices, reducing the load of energy-consumption peaks, facilitating both short-term and long-term prediction, eliminating blackouts, and promoting the use of local and greener energy resources. Consumers will be able to respond to price changes by reducing or shifting their consumption to low-price slots [3] but also by informing energy providers of their future schedules. For example, after an increase in electricity prices during a hot summer week, consumers could accept a lower tariff to move a portion of their consumption load to low-peak hours. They could also provide information about their schedules, e.g., a planned absence for vacation, helping energy providers to improve prediction.

On the other hand, some early studies [4] have shown that although people generally show interest in environmental issues, they are less willing to spend time making decisions about their daily energy use. The goal of this internship is to further understand such attitudes and come up with solutions that maximize the overall utility of negotiations by reducing the participation effort of energy users. We are especially interested in investigating viable negotiation protocols and designing mobile user interfaces that minimize the overhead of decision-making.

The student is expected to:

  1. Understand and review negotiation protocols and optimization strategies for the smart grid proposed by previous research.
  2. Conduct an early survey that explores people’s attitudes towards residential energy consumption. The survey will focus on how people view consumers’ active participation given different negotiation protocols and the role of interactive technology in decision-making.
  3. Explore the design of a mobile interface that combines direct interaction with automated support to help users make quick decisions on their schedules and agree on effective energy-consumption shifts.

Given enough time, the student could also work on the design of a user study that observes the use of the mobile interface under a semi-realistic setting, where demand and pricing offers are determined through a controlled simulation.

The internship can be 4 to 6 months long.

Required skills

We are looking for Master students who are enthusiastic about user-interface design and research in Human-Computer Interaction but are also interested in aspects of Artificial Intelligence and decision theory. Solid programming skills are required, preferably in Java. Experience with the Android platform will be a plus.


  1. Ramchurn, S. D., Vytelingum, P., Rogers, A, and Jennings, N. Putting the 'Smarts' into the Smart Grid: A Grand Challenge for Artificial Intelligence. Communications of the ACM, pp. 86-97, 2012.
  2. Linky, le compteur nouvelle génération.
  3. Akasiadis C. and Chalkiadakis G. Agent Cooperatives for Effective Power Consumption Shifting. In Proceedings of AAAI, pp. 1263-1269, 2013.
  4. Rodden, T. A., Fischer, J. E., Pantidi, N., Bachour, K., and Moran, S. At home with agents: exploring attitudes towards future smart energy infrastructures. In Proceedings of ACM CHI, pp. 1173-1182, 2013.