**1. Introduction**

Energy systems worldwide are undergoing a radical transition to low-carbon energy sources. This transition is necessary for countries to achieve their nationally determined contributions (NDCs) as per the Paris Agreement of 2015. The International Renewable Energy Association (IRENA) global roadmap for energy transformation, ReMap, has quantified that, for countries to achieve their NDCs, renewable energy sources should account for two-thirds of the total primary energy supply worldwide by 2050 [1].

ReMap also calls for large-scale electrification of the energy demand. Currently, electricity accounts for 20% of the final energy demand worldwide, according to the roadmap this ought to be 49% by 2050. Therefore, to meet the intended NDCs, considerable electrification of the final energy demand and a tripling of the installed capacity of renewable electricity sources, when compared to its current levels, should occur simultaneously around the globe. Additionally, due to their extended availability and continuously reducing costs, variable renewable energy sources (VRE), particularly wind and solar energy, are expected to be the primary sources of 61% of the total electricity generated worldwide [2].

Under this scenario, electrical grids worldwide will be subject to extreme stress on two fronts, a vertiginous growth in demand, while primary energy sources are substituted by considerably volatile replacements. Consequently, for the energy transition to be successful, grid operators have to be equipped with new grid management opportunities. Hitherto, these operators have relied on a combination of base and peak load power plants that adapt their output to balance changes in demand. Nonetheless, in an electrical grid that relies on a high share of VREs, new forms of grid flexibility are essential. Grid flexibility allows the balance of both end sides of the electrical grid, supply and demand, hence ensuring grid stability.

Grid flexibility options in a VRE-centered electrical grid include supply-side energy flexibility (SSEF), storage at grid level, grid expansion and demand-side energy flexibility (DSEF) [3]. SSEF consists of the diversification of primary energy sources, increasing the share of dispatchable sources, which can adapt their electrical output to offset any unexpected output deviation of VREs. Energy storage consists of a series of facilities connected directly to the grid with the sole purpose of storing different energy forms that act as a buffer between electrical supply and demand. Grid expansion goes in tandem with energy storage and involves the expansion of high-performance electricity grids, which can transport and distribute the electricity over wide territorial spaces, aggregating VREs with different generation profiles. All of these options involve a considerable additional investment in infrastructure that increases electricity costs and in some cases, might even involve the reliance on non-renewable energy sources [4].

The final option, DSEF, comprises the capacity of the demand sectors within the electrical grid to adapt (increase, reduce or shift) their electrical consumption over a specific duration to balance variations in the electrical supply [4,5].

Among the sectors that constitute the electrical demand, DSEF of the industrial sector or industrial energy flexibility (IEF) is of particular interest. From the perspective of the grid operators, the share size of the industrial sector in the electrical demand, which in the EU-28 represented 37.4% of the electrical consumption in 2017, and as mentioned is expected to grow, makes it a prime candidate to add flexibility to the electrical grid [6]. From the companies perspective, the high relative electricity costs when compared to the cost of other energy carriers, i.e., natural gas, added to the increased control in their energy consumption makes IEF an attractive optimization opportunity [7,8]. In contrast to SSEF, storage at grid-level and grid expansion, DSEF, in general, and IEF, in particular, allow the techno-economic optimization of energy consumption, potentially reducing instead of increasing the overall energy, predominantly electricity, costs for the industrial sector [9].

The tracking report of the International Energy Agency (IEA) in the topic of Demand Response (DR), one of the applications of DSEF, shows that, despite the above-mentioned benefits, the industrial sector still plays a minor role in the current and expected DR potential [10]. As the results of the industRE and SynErgie projects show, this in part due to a lack of knowledge from companies of the energy flexibility capabilities of their production facilities and of the prospective benefits they can obtain by exploiting these capabilities [8,11]. Hence, there is a need for third-party industrial energy audits that support the industrial sector towards systematically identifying and quantifying the energy flexibility capabilities of their facilities and estimating the associated benefits of exploiting such capabilities.

This article proposes an answer to this problem through a systematic, industrial system-focused methodology. The methodology consists of delimiting and classifying the different industrial systems that constitute a production facility, establishing which systems are suitable for flexible operation and, identifying and characterizing the available energy flexibility measures in those systems deemed suitable. The characterization includes a calculation of the potential economic benefit from the usage of these measures. The proposed methodology combines the existing practices of industrial energy auditing with the state of research in the topic of industrial energy flexibility through an innovative approach that complements the current widespread focus in industrial energy efficiency.

The article is structured as follows. The understanding of production facilities from an energy perspective is described in Section 2. Section 3 defines the key concepts of DSEF and IEF. In Section 4, the proposed methodology to systematically identify and characterize energy flexibility measures is explained in detail. Section 5 illustrates the application of the methodology by summarizing its application in an existing industrial system, a chilled-water air conditioning system. Section 6 discusses the proposed methodology and the results of its initial application. The paper concludes with several final insights and an outlook of the prospective applications of the proposed method.
