The role of AI and advanced analytics in mitigating the impact of the Energy Crisis

Updated: Apr 27

After more than two years of the Covid-19 pandemic, 2022 was optimistically projected as the starting point of the post-covid era. However, in late 2021 a number of organizations (e.g. OECD) had already suggested that 2022 could bring together the "ingredients" of a "perfect storm" for the Northern Hemisphere economies: high energy and commodity prices; cyberattacks; rising labour costs; difficulty in retaining and hiring talent; pressure on supply chains; political deadlock in the US; and geopolitical tensions, particularly on the Ukrainian-Russian border.

The year is not yet half over and we can see that, in fact, 2022 has brought companies much more than this imperfectly forecasted "perfect storm". 2022 "woke up" to an unimaginable war, characterized by a barbarity that we thought was buried with the Second World War.

The conflict is intensifying the escalation in the prices of oil, natural gas and electricity, which were already rising in 2021. Given the European Union's sanctions on Moscow and the growing political, social and media pressure to eliminate the import of oil and natural gas from Russia, the projected cost of energy in 2022 will be increased by the cost associated to the embargo on Moscow.

For companies, this geopolitical context means that energy, as a productive factor, will have an increasing cost, in many cases beyond the sustainability threshold of their business. Considering that energy prices are expected to decrease or stabilize - on the contrary - how can this negative impact of the cost of energy be mitigated? Even without a "magic wand", the adoption of strategies based on advanced data analytics and artificial intelligence (AI) models makes the difference in the competitiveness of companies facing the energy crisis.

Firstly, a natural gas-intensive company (e.g. steel, ceramics, paper, glass companies) should evaluate the possibility of implementing fuel shifting. This evaluation, based on advanced analytical models and rooted on internal and exogenous data, will necessarily include the analysis of the consumption profiles of industrial units and the type of technologies available in the market, in order to promote the replacement of natural gas consumption by electricity (desirably in a self-consumption regime) or by green hydrogen (i.e., generated from renewable energy sources).

In the specific case of electro-intensive consumers, particularly affected by escalating electricity prices, the recent approval of the exemption of the application of the proximity criteria between production units and the location of the facilities has decisively opened the door of renewable energy production for self-consumption. AI models may assist in forecasting renewable energy sources, as well as optimizing energy balance (between renewable energy production and electricity consumption in the industrial unit), including, for example, energy storage technologies.

In parallel, the implementation of Renewable Energy Communities presents itself as a prime alternative for many companies/business groups. In this context, a well-developed AI algorithm adjusted to the needs of all members of the community will maximize endogenous renewable resources (through optimized consumption management), minimizing the amount and cost with energy purchased from the grid, and maximizing the remuneration with energy trading operations and flexibility services.


On the other hand, in addition to energy consumption related to industrial facilities, it is important to address the fuel demand of the company's vehicle fleet. If electric vehicles are indeed a cost-efficient option for urban mobility, the vast majority of corporate fleets are composed of heavy (goods) vehicles (HGV) and light vehicles covering daily distances incompatible with electric mobility.

Optimal route management is the most effective strategy to promote fleet operational efficiency. Optimal fleet routing models allow companies to minimize distance travelled and the fleet consumption, and in some cases, even to resize the capacity of the vehicles, always maintaining the quality of the service.

It is nothing new to say that Big Data and advanced analytical models are here to stay as tools to respond to the challenges faced by companies in a global world. However, in a context of energy crisis like the one we are now experiencing, in which traditional response mechanisms do not offer viable alternatives, this reality is inescapable.

Miguel Moreira da Silva


Article originally published in "Jornal Económico" (Portuguese only)