The implementation of the iFLEX solutions in the three pilot sites: Finland, Greece and Slovenia has led to a set of key lessons learned in terms of technology, user engagement and economy.

The technical dimension


  • The wide rollout of smart meters is crucial, as it will enable collecting higher quality data (e.g. eliminating downtime), which will also be validated by the Distribution System Operator, from a large number of end users. This can serve as a base for the development of improved accuracy models for tools such as Digital Twins. Furthermore, it could result in estimating the flexibility potential of various types of end users, including residential ones in an efficient way.
  • The installation of smart HEMS systems that are able to connect to already installed devices at the end-user premises via wireless network and without interference with concealed electrical installations is extremely simple, fast and efficient procedure, as a number of measuring and control parameters can be obtained from the devices via already existing communication protocols.
  • Heterogeneity of Building Automation System interfaces is a challenge as regards their interoperability with the Resource Abstraction Interface module of the iFLEX framework, which is expected to be addressed in the future with the implementation of standards like SAREF4ENER and EEBUS.
  • Hybrid modelling is a good solution for the HVAC systems, since accurate and robust models can be created without a need to physically model all the details of the HVAC system. This will allow for the replication of the solutions more easily for different types of buildings and heating systems. Further work is needed to develop a more robust model (i.e, not a pure machine learning model), which however can be replicated to different types of buildings without any manual modifications needed (i.e., all the differences are learned from data).
  • Following the domain standards, the interactions with the Demand ResponseDemand Response Intentional change of normal consumption patterns by energy consumers in response to external price signals (implicit DR) or incentives to reduce or increase consumption (explicit DR). Both operations can be automated. Management System (DRMS) were modelled on the basis of OpenADR2.0 profile (also available as IEC 62746-10-1 ED1). However, the use of this protocol could prove a bottleneck in the design, since it was designed for field device communication, rather than cloud-based communication as it is in the cases of many iFLEX Assistant components relevant in the DRDR Demand Response communication lifecycle. A recommendation would be to extend OpenADR for cloud-based communications or DR clients (also known as VENs) managing multiple customers.
  • Security and privacy can be provided with a very light overhead and access to the iFLEX Assistant resources can be provided in a fine grain manner. Privacy requirements can be fulfilled even if the RAI backend systems are provided in the cloud.


  • Heating systems constitute a fair source of flexibility both in terms of power and energy. This is because most of the energy demand in buildings comes from heating, and building thermal mass can be used for storing heat (energy) for several hours.
  • Automation of the flexible assets, based on the end users’ preferences, greatly facilitates the provision of flexibility, as it does not require the physical presence of the end user at the premises to activate or deactivate manually a device. On top of that, automated operation enables avoiding end-users’ fatigue, which is caused by multiple interactions with energy and flexibility management applications, such as iFLEX Assistant.
  • In some pilot sites, such as Greece, the flexibility aggregation from residential end users would benefit from a larger and more diverse portfolio of flexible assets (e.g. EVs and heat pumps).
  • The model-based approach provided by the Digital Twin repository and the Automated Flexibility Management module works well for optimal control and makes it possible to estimate the baseline load (power forecast) and flexibility, as described in the S2 interface standard.

User interfaces

  • The agile approach that was followed for the design and implementation of the iFLEX app, which aimed at developing a user-friendly interface, was found to be efficient and produced high-quality end results.

User engagement and acceptance dimension


  • Easiness in recruiting potential end users is very important to simplify the recruitment procedure. This entails concise and informative project presentation, effortless signing of associated informed consent forms and clear presentation of potential benefits for end users.
  • A clear pilot user journey and early pilot user involvement provide a lot of benefits, both in terms of user satisfaction and the sense of ownership as well as easier adaptability and less risk of expectation misalignment.
  • Although potential pilot participants show very strong interest in participating, various restrictions may reduce the number of eligible pilot users and/or their enthusiasm significantly. Among others, the unavailability of flexible electrical loads (e.g. electric water heater) at their premises, the unavailability or poor quality of telecommunication infrastructure for the unobstructed communication of the installed equipment or the fact that some electrical boards do not fulfil minimum requirements for the installation of the necessary technical equipment (e.g. aged boards in poor condition, not enough space for additional devices) may hinder the flawless installation, testing and validation procedures.
  • In general, it is beneficial for the flexibility procurer to select end users that are more responsive to incentives, because the offered DR incentives will be smaller. In this context, the total flexibility earned will be maximised when selecting more responsive users that also require a small amount of minimum incentives to participate. Adequate knowledge to discriminate users in groups (according to their responsiveness to incentives), transfers part of the profit from the users to the provider without necessarily affecting the flexibility target. The provider should employ clustering techniques to split the user base in groups of similar characteristics. This will magnify the revenue and subsequently profit without increasing the budget.


  • Since end consumers may be generally reluctant to allow for even minor interventions at their premises (e.g. installation of new hardware in the electrical boards, the need to maintain reliable operation of internet connection to allow for uninterrupted communication of all devices and software platforms, personal data protection concerns, etc.), a clear and prompt communication to pilot users is of utmost importance. Or else, the latter are faced with too many unknown aspects, including new technologies (i.e. technical equipment installed, new web/mobile applications), new business models and use cases (i.e. adjusting energy consumption according to price or network signals, need to provide instant feedback, accept/reject energy advice, operation of incentive mechanisms), new companies and institutions (i.e. project stakeholders, market authorities, institutions and operators, project partners, equipment vendors and subcontractors), new data streams to be shared (i.e. personal data to be shared, energy consumption data, personal preferences, habits).
  • Compliance with GDPR is paramount. In particular, residents that are interested in having equipment installed should register and provide their consent for data collection. Depending on the local legislative and regulatory framework, several legal complexities and implications may arise as regards the person(s) that should sign the associated consent form(s) i.e., the legal owner of the property or pilot end user(s). In the same context, clauses regarding fair usage of the smart metering equipment, prohibiting unauthorised users from tampering with the installed equipment should be clearly included in the consent form. Finally, the equipment (and particularly any sensors that collect personal data) are resident/household-specific and, therefore, when a new resident moves into an apartment, a new personal data collection agreement must be approved/signed or data collection should be stopped.


  • Web/mobile interfaces designed for residents and property managers, which can be used to: a) monitor total energy consumption, environmental impact and residents’ apartment-specific data, b) provide energy advice, d) facilitate the involvement of the end consumers in DR actions through push notifications or other mechanisms, and d) visualise concrete benefits from their active participation in the project, are fundamental tools for the fulfilment of the project objectives. Innovative functionalities, such as the possibility for the end user to provide feedback immediately if the living conditions become worse due to controls in the pilot, assist towards maximising user acceptance and engagement.
  • Feedback surveys were also found to be a good means of collecting user experiences about the project and to detect changes in the end consumers’ awareness and consumption behaviour. Suggestions to improve the user interface as well as requests for the provision of additional data need to be addressed to keep the end users’ interest unabated.

The economic dimension


  • Monetary incentives are naturally in the interest of the building owners/end consumers and they usually depend on the type of the end consumer:
    • For small electricity consumers (e.g. households) where no spot price-based contracts are available yet, any cost reductions can mainly be obtained through energy efficiency (i.e. reduction of total electricity consumption) and by optimally selecting the most efficient heating source (e.g., heat pump or district heating).
    • In the case of large commercial buildings, where spot-price based contracts can be more easily adopted, demand-side flexibility can be used for load-shifting of flexible loads (i.e. shift the electricity consumption from high market price hours to time intervals where lower market prices apply). In this case, how the flexibility will be used has to be forecasted on a day-ahead basis, as the total volume of electricity needed is mostly purchased from the day-ahead market.
  • It may be almost impossible to alter the billing systems of the electricity supply companies involved in the project to apply customised discounts or any special tariffs related to the incentive policies on end consumers. Unfortunately, in some pilot sites such suggestions are currently not feasible from a regulatory point of view, without a clear estimate on when it would be possible to facilitate this type of transactions.
  • Exogenous rewards based on an agreed reward mechanism that is not linked to electricity usage or billing but is still designed to motivate users’ enthusiastic participation in innovative DR projects could be a viable alternative. However, piloting showed that high-value prizes (i.e. a new mobile phone) may have no impact to the residents’ activity to participate and only those residents that are genuinely interested in energy awareness and energy flexibility participate actively in the pilot. In other cases, direct monetary rewards in the form of redeemable coupons are able to boost the participation of end users in relevant DR actions.
  • Non-monetary incentives to end consumers can also be successful: Energy awareness and particularly the increase of awareness regarding the effects of each resident’s own energy consumption behaviour, the familiarisation of residents with the concept of energy flexibility and the ways in which a household consumer can provide flexibility are decisive factors towards promoting their participation in energy flexibility actions in the future.

Business opportunities

  • Heating systems are a great asset for demand-side flexibility and provide significant monetary benefits. The monetary benefits are naturally higher in case of spot price-based contracts for electricity consumption.
  • Reselling by the provider of the harnessed flexibility in the market gives rise to a different value chain that offers new opportunities and indirectly associates the grid benefit to stronger user and provider incentives. With careful selection of parameters, i.e., incentives budget and flexibility unit resale price (if negotiable) can lead to “all-win” situations that is beneficial for all players involved in the corresponding value chain.
  • As regards the implementation of a novel bilateral trading scheme between a RES and DR AggregatorAggregator Market participant acting between providers and users of flexibility. The Aggregator bundles flexibility resources (consumption, generation, storage) into a flexibility volume and offers the capability to activate it. for mitigating imbalances in RES generation within an augmented common portfolio of RES units and end-user demand-side flexibility resources, using DR resources has proven to be beneficial for all the three types of stakeholders involved (i.e. RES Aggregator, DR Aggregator, end consumers). However, dividing the profits among the three stakeholders can be a complex exercise where everyone will want to maximise own profit. If this is not done carefully, taking into consideration the proper incentivisation of the other two entities, the program’s effectiveness might be jeopardised. The prudent definition of the bilateral contract price between RES and DR Aggregators is an important part of the bilateral trading process. Very low contract price values do not provide any profit for the DR Aggregator and a fair starting point would be any value that is close to the day-ahead market clearing price. Finally, both Aggregators and end users have more room for profit in the case of a large RES generation imbalance.