Modeling and Optimizing the Process of Identifying Energy-Saving Potential Scope (ESPS) in Municipalities: A Combinatorial Approach to ISO 50001 Implementation

Abstract

The energy consumption of buildings, the effectiveness of energy-saving measures, and the exploitation of energy-saving potential are strategic issues for improving the energy performance of public assets and limiting their environmental impact. However, small and medium municipalities (SMMs) encounter difficulties in identifying their energy-intensive units, a process that is often lengthy (3 to 18 months), costly, and dependent on traditional methods such as the Real Estate and Energy Master Plan (REMP) promoted by ADEME or the Cit’ergie system. These approaches, although structured, rely on time-consuming manual analyses that require significant technical and human resources. This article proposes an innovative solution, PG2E, based on a combinatorial approach that quickly identifies Energy-Saving Potential Scope (ESPSs) from energy consumption data. Backed by a realistic–critical stance to assess the limitations of existing systems and a constructivist–pragmatic approach to designing a tool adapted to SMMs, the PG2E solution uses simple statistical criteria (average, upper quartile). This study, conducted in the town of Quesnoy-Sur-Deûle, shows that PG2E identifies ESPS with a success rate of 60% to 100% while reducing time and costs. It thus offers an accessible digital alternative for initiating an approach that complies with the ISO 50001 standard.

1. Introduction

The global energy challenge is central to the climate change debate due to our historical reliance on fossil fuels and their significant contribution to greenhouse gas (GHG) emissions. The 1.5 °C limit set out in the Paris Agreement is an ambitious target that will require far-reaching transformations in energy, economic, and social systems [1].

The scientific literature converges on the fact that limiting global warming to 1.5 °C requires rapid, systemic transformations in all economic sectors [2]. The scientific foundations laid down by the IPCC (Intergovernmental Panel on Climate Change) call for urgent action to reduce global emissions as early as this decade [2]. The trajectories proposed by Rogelj et al. [3], as well as the energy scenarios drawn up by the IEA (International Energy Agency) [4] and Grubler et al. [5], offer strategic frameworks to guide this transition. However, as highlighted by the UN [6], there remains a significant gap between stated ambitions and actual implementation. Bridging this gap is essential to avoid irreversible climate impacts while ensuring equitable sustainable development.

Among the sectors requiring far-reaching energy transformation, the building industry occupies a central position due to its major role in energy consumption and greenhouse gas emissions.

The energy efficiency of buildings is a central issue in energy transition and climate change policies. With an estimated contribution to global energy consumption in developed countries of between 20% and 40%, the building sector represents an essential lever for reducing greenhouse gas emissions [7]. Several studies have shown that heating, ventilation, and air-conditioning (HVAC) systems are the most energy-intensive consumption items, accounting for almost 50% of total building energy consumption. However, despite the existence of numerous energy efficiency policies, energy consumption continues to rise, particularly in the tertiary and residential sectors, underlining the importance of developing tools that are better suited to identifying potential energy savings.

The literature used (

Table 1. Comparison of main contributions focused on energy consumption.

References	Main Objective	Methodology	Main Contribution
[7]	Analysis of building energy consumption.	Data review and cross-country comparison.	Highlighting the growing weight of the building sector in global consumption, particularly through HVAC systems.
[8]	Modeling the energy consumption of machine tools.	Classification of energy consumption models.	Identification of the three main categories of energy models applied to industrial processes.
[9]	Development of an energy consumption allowance (ECA).	Segmentation of energy consumption steps (ECSs).	Proposal of a modular approach to quantify and optimize energy consumption in manufacturing systems.

) has mainly focused on building energy consumption and energy-saving prediction models [8]. One of the most common approaches relies on a detailed analysis of the energy flows and equipment in place, enabling energy-saving potential to be identified through advanced simulations and detailed energy audits. Although robust, these methods generally require substantial resources in terms of technical expertise, data collection and financial resources. In addition, a large body of research has examined energy efficiency measures, from technological improvements (equipment refurbishment, thermal insulation, etc.) to more effective energy management strategies (optimizing usage, modifying user behavior) [9].

In the face of ever-increasing energy consumption in buildings, research has also focused on the potential for energy savings. Drawing on the sources presented in

Table 2. Comparison of main contributions based on energy-saving potential.

References	Main Objective	Methodology	Main Contribution
[10]	Provide an overview of electricity consumption and efficiency trends in the enlarged EU and candidate countries, as well as the current market share of efficient appliances and equipment. Estimate the potential for electricity savings in buildings.	Analysis of existing data on electricity consumption, market share of efficient equipment, and energy policies.	Provides a baseline for electricity consumption and energy efficiency in the enlarged EU. Identifies key areas for energy savings.
[11]	Assess the mitigation potential of greenhouse gas (GHG) emissions in residential and commercial buildings.	Review of over 80 recent studies from 36 countries and 11 country groups.	Provides a comprehensive assessment of the potential for mitigating GHG emissions in the building sector, as well as the associated costs. Highlights the barriers to realizing this potential and the policies that can be used to overcome them.
[12]	Examine the economic concepts underlying consumer decision-making on energy efficiency and conservation, and review the related empirical literature. Assess the extent to which these conditions motivate policy intervention in energy-using product markets.	Review of the economic literature on market barriers, market failures, and behavioral failures that have been cited in the context of energy efficiency.	Provides an economic perspective on why energy efficiency policies can be justified. Highlights gaps in knowledge about the extent of certain market and behavioral failures.
[13]	Examine the factors influencing Greek households’ intentions to adopt energy-saving measures.	Survey of 586 Greek households. Econometric analysis of survey data.	Identifies the economic, demographic, housing-related, and attitudinal factors that influence the energy-saving intentions of Greek households.
[14]	Examine five possible approaches to assessing energy savings, which could form the basis of an international agreement.	Examination of the strengths and weaknesses of each approach and discussion of the lessons learned from carrying out this evaluation process.	Provides an overview of the different approaches to assessing energy savings and their implications for the establishment of an international GHG emissions trading system.

, the literature review aims to identify the general trends, the challenges encountered, and the policies put in place to improve energy efficiency.

Despite numerous policies and initiatives at the European and national level, electricity consumption continues to rise in EU member countries. In the EU-25 residential sector, for example, consumption rose by 10.8% between 1999 and 2004, keeping pace with economic growth. Comparable trends were observed in the tertiary sector, although the increase was less marked in industry. However, there is still a lack of precise data on the breakdown of this consumption between different uses and types of equipment, particularly in the tertiary sector, where information is even more limited than in the residential sector [10].

The European Union has introduced a number of energy efficiency programs and policies to encourage reduced consumption, including standards, energy labels, financial incentives, and energy-saving certificates. One of the most effective levers for improving energy efficiency is to speed up the replacement of energy-hungry equipment with more efficient models, thus reducing the time needed to completely renew the existing stock [10]. The existing challenges can be summarized as follows:

• Cost of data collection: identifying potential energy savings requires in-depth analysis, which is often costly and difficult to implement [11].
• Lack of appropriate financial incentives: the lack of coordination between the various players (e.g., the owners who make the investments and the tenants who benefit from them) limits the implementation of effective measures [11].
• Fragmentation of the building sector: the diversity of the players involved (architects, engineers, construction companies, operators) makes it difficult to take a global, coherent approach to energy efficiency [11].

These challenges are particularly acute in the residential and tertiary sectors, which calls for a more integrated approach combining several policy tools and instruments.

While identifying energy-saving potential is an essential first step, implementing concrete measures remains a major challenge, according to the articles sourced from the energy-saving measurement literature presented in

Table 3. Comparison of main contributions based on the energy-saving measure.

References	Main Objective	Methodology	Main Contribution
[15]	Examine energy efficiency policies in OECD countries.	Analysis of energy intensity trends.	Identifying effective policies and key factors for improving energy efficiency.
[16]	Analyze bottom-up models to assess energy consumption in the residential sector.	Critical review of bottom-up physical models; comparison of five models for the UK.	Highlighting the strengths and weaknesses of different models and identifying future modeling needs.
[17]	Analyze the gap between actual and optimal energy use.	Distinguishing between different notions of optimality (economic potential, social optimum).	Identification of market failures and non-market failures as causes of the gap.
[18]	Combine bottom-up and top-down approaches in an energy demand model.	Estimation of model parameters from bottom-up information, taking into account the energy efficiency gap.	Proposal of a model that reconciles bottom-up and top-down approaches; improved policy analysis.
[19]	Examine the co-benefits of improving energy efficiency in buildings.	Review of the literature on quantifying co-benefits and proposal of a methodology for incorporating them into cost–benefit analyses.	Highlighting the importance of co-benefits and proposing a methodology for quantifying them and integrating them into decision-making analyses.

.

Improving energy efficiency relies on a combination of policies, assessment models, and analyses of consumption behavior. These different approaches enable us to identify the most effective strategies for reducing energy consumption and optimizing the energy performance of buildings.

Energy efficiency policies play a key role in reducing energy intensity. Geller et al. (OECD) analyzed policies implemented over a thirty-year period, highlighting their impact on reducing energy consumption. Their results show that these gains are often attributed to changes in economic sectors and technological innovation but that research and development (R&D) efforts in the building sector remain under-exploited [15].

Different methodologies are used to assess energy consumption and identify opportunities for improvement. Kavgic et al. have investigated bottom-up models, which estimate baseline building energy consumption and analyze the impact of CO₂ emission reduction strategies. However, their study highlights several limitations, including the lack of detailed public data and the absence of models incorporating both technical and socio-economic factors [16].

Jaffe and Stavins have explored the difference between actual and optimal energy use. Their work highlights several market failures, such as the lack of information on potential energy efficiency gains or the high initial cost of retrofits, which prevent widespread adoption of energy-saving solutions. They distinguish several levels of optimality, ranging from theoretical economic potential to the social optimum, depending on incentives and policy interventions [17].

Koopmans and et Velde have proposed a hybrid model, combining bottom-up and top-down approaches to predict energy demand. Their study reveals a slowdown in the adoption of energy-saving technologies between 1990 and 2015, suggesting that more incentive-based policies and accessible innovations are needed to accelerate the energy transition [18].

Beyond simply reducing consumption, Sharmina et al. have identified several co-benefits linked to building energy efficiency, such as the following:

• Improved indoor air quality.
• Reduced respiratory illness and mortality.
• Productivity gains and enhanced well-being for occupants. However, these benefits are rarely integrated into economic analyses and political decisions, which underestimates their overall impact [19].

Research into energy-saving measures faces several challenges:

• Lack of data transparency and accessibility: Gathering reliable information on consumption and potential savings remains a major obstacle [16].
• Limited consideration of socio-technical factors: Studies do not always take into account occupant behavior and its impact on energy consumption [16].
• Lack of standardized methodologies for quantifying and comparing the co-benefits of energy savings [19].
• Multidisciplinary, dynamic models integrating energy, economics, and public health for a more global vision of energy efficiency are needed [16].

Energy-saving measures are essential for reducing consumption and improving the energy performance of buildings. However, efforts are still needed to refine evaluation models, integrate co-benefits into decision-making, and harmonize methodologies. Implementing more accessible solutions, tailored to the realities of public building and businesses, is a key issue in maximizing the long-term impact of energy efficiency policies.

An analysis of the challenges and gaps in the existing literature on building energy consumption, energy-saving potential, and energy-saving measures highlights the lack of solutions tailored to small and medium municipalities (SMMs) for identifying energy-saving. This limits their ability to optimize their efforts to reduce consumption and comply with environmental standards, notably ISO 14001 and, more specifically, ISO 50001 for energy management systems (EMSs) [20]. The development of methodologies and tools dedicated to SMMs would help fill this gap, facilitating the identification of energy-saving potential scopes and reinforcing the effective implementation of local energy policies.

The main objectives of this article are to

• Analyze and assess the limitations of approaches current to identifying energy-saving potential, highlighting the specific obstacles encountered by small and medium municipalities (SMMs).
• Develop an innovative methodology, based on combinatorial optimization and simplified criteria, to enable a rapid and effective identification of potential energy savings.
• To propose an automated digital tool, designed to facilitate the identification of energy-saving potential for SMMs and enable them, without advanced expertise, to determine the perimeters eligible for energy management system (EMS) certification under the ISO 50001 standard. This tool will be tested through a case study.

By combining an adapted methodological approach and an accessible digital tool, this research aims to offer SMMs a fast, reliable, and low-cost solution to optimize the process of identifying energy-saving potential and strengthen their commitment to the energy transition. Before exploring these solutions, it is essential to precisely define the concept of ESPS and to examine existing approaches to identifying them in the specific context of SMMs.

2. Energy-Saving Potential Scope (ESPS)

2.1. What Is the ESPS

The energy-saving potential scope (ESPS) is defined as a set of energy-intensive units, zones, or sectors where energy savings can be achieved. Figure 1 shows an example of an organization’s ESPS, with five energy-intensive units. The aim is to identify the scope where energy consumption can be reduced without affecting service or production levels. Energy-saving opportunities can be found in various sectors, such as industry, construction, transport, and agriculture. They include actions such as improving the energy efficiency of equipment, reducing energy losses, and adopting more energy-efficient behavior [21,22].

The process of identifying the energy-saving potential scope (ESPS) depends on the system used in the municipalities. And the ISO 50001 certification scope (CS) is derived from the relevant ESPS.

This concept is fundamental to assessing and improving energy efficiency in different contexts, and understanding it is essential to any research aimed at optimizing the management of the process of identifying energy-saving opportunities.

Based on set theory, we identify the subsets of a set of five energy-consuming units, for example, excluding the empty set (∅) and the full set, and calculate their count, which is 2⁵ − 2 = 30 subsets grouped by subset type characterized by its count. In our example, we obtain the following types of perimeter or subset in

Table 4. Example of perimeter count for a set of 5 energy consumption units.

Pn	Scope Description	Quantity Pn
P1:	Energy-saving potential scope set with 1 energy-intensive unit	5
P2:	Energy-saving potential scope set with 2 energy-intensive units	10
P3:	Energy-saving potential scope set with 3 energy-intensive units	10
P4:	Energy-saving potential scope set with 4 energy-intensive units	5

:

2.2. Current Methods Used to Identify the ESPS

Energy data collection methods for identifying ESPS in municipalities are generally based on several existing solutions, including the local measurement system (LMS), the Real Estate and Energy Master Plan (REMP), and the Cit’ergie System (climate–air–energy policy labeling program), according to the field survey (Figure 2).

The survey shows that current methods for collecting data and identifying ESPS are poorly adopted, with the exception of the local measurement system (LMS). It also examined the criteria used by municipalities to identify the ESPS (Figure 3).

The results of the survey show that the physical obsolescence index (POI) of units, along with other criteria such as regulatory status used by municipalities, account for 54% of energy-hungry units.

An organization’s or municipality’s energy policy defines the criteria for existing systems, among which it is essential to select the most relevant ones in order to identify the optimal ESPS and define it as a certification scope (CS) according to the ISO 50001 standard. Criteria may include the following:

• Energy consumption of the energy-hungry unit.
• Energy consumption per square meter of the energy-consuming unit, if it is a building.
• Energy Performance Diagnostic (EPD).
• The indicator that qualifies the state of disrepair of the building, equipment and networks.
• Regulatory status.
• The cost of owning the building for the community.
• Etc.

The classic methodology of the current process used to identify energy-saving potential scope (ESPS) is a step-by-step process (Figure 4):

A walk-through audit can be conducted as follows:

• Visual Inspection: A rapid inspection of installations to identify energy-saving opportunities linked to equipment maintenance and operation.
• Rapid Identification: This quickly identifies scopes with high saving potential without a detailed analysis [23].

A Standard Energy Audit can be conducted as follows:

• Data Collection: Detailed collection of energy consumption data over a significant period.
• Consumption analysis: Evaluation of energy consumption by use (heating, lighting, etc.) and identification of inefficiencies.
• Recommendations: Proposed energy efficiency improvement measures with estimated potential savings and return on investment [23].

An in-depth energy audit can be conducted as follows:

• Energy Analysis: Assessment of the quality of energy used and identification of exergy losses for specific processes.
• Modeling and Simulation: Using thermodynamic models to simulate energy performance and identify energy-saving opportunities [23].

The existing systems for identifying energy-saving potential scope (ESPS) are as follows:

• The Real Estate and Energy Master Plan (REMP) method.
• The local measurement system (LMS).
• The Cit’ergie system.

In the following section, we present these different methods and their application in detail in the context of small and medium municipalities (SMMs).

2.3. REMP System Method for Identifying the ESPS

The Real Estate and Energy Master Plan (REMP) is a strategic management tool designed to help municipalities control their energy consumption and optimize the management of their built heritage. Its main objective is to plan, within the framework of the Pluriannual Investment Plan (PIP) [24], priority energy improvement actions, particularly in response to regulatory requirements such as the Eco Energy Tertiary Decree [24,25]. To identify energy-saving potential scope (ESPS), the REMP method combines complementary approaches such as building mapping [26], energy audits, multi-criteria analysis (taking into account age, usage, energy performance, etc.), and strategic planning [24] (Figure 5).

This approach, although rigorous and suitable for comprehensive management, has significant limitations, particularly for small and medium municipalities (SMMs): it requires specialized engineering, mobilizes significant human and financial resources, and can take 3 to 18 months to produce operational results. As a result, many SMMs are unable to effectively implement a comprehensive REMP, which hinders their ability to initiate an energy management system (EMS) compliant with ISO 50001 within a reasonable timeframe and with controlled resources.

2.4. Measurement System Method to Identify the ESPS

The energy measurement system comprises all the devices used to measure, record, and analyze the energy consumption data of a building or site. It is based on three main components: measurement devices (sensors, meters), recording devices (data acquisition chains), and analysis devices (data exploitation tools) [27,28,29] (Figure 6).

This approach makes it possible to centralize consumption information (electricity, gas, water, etc.) and analyze it to model energy performance. The objective is to monitor consumption in real time, identify deviations, calculate energy performance indicators (EPIs), and detect opportunities for improvement [27,29,30,31]. Recommended under the ISO 50001 standard, this system enables better control of energy use by supporting the planning and management of optimization measures.

The data collected by a local measurement system are analyzed to identify energy savings through several methodical steps, as shown in Figure 7:

However, using the measurement system to identify energy-saving potential scope (ESPS) in small and medium-sized communities (SMMs) poses several challenges. These systems rely on manual or semi-automated analysis, often using Excel, making the process long and complex. Their implementation requires significant investment in equipment, software, and technical skills, which is often beyond the reach of SMM [32]. In addition, managing massive volumes of data requires advanced storage and analysis infrastructure. Finally, interpreting the results to detect ESPS requires specialized expertise, which is difficult to mobilize in structures with limited human and financial resources. These constraints limit the accessibility and effectiveness of this method for municipalities with low organizational capacities [33,34,35,36].

2.5. Cit’ergie System Method for Identifying the ESPS

The Cit’ergie system is a management and certification program supported by ADEME, aimed at helping local authorities develop and implement ambitious policies on energy, climate, and air quality. Structured around the “Territorial Climate-Air-Energy Plan” (TCAEP), Cit’ergie helps local authorities to structure, monitor, and evaluate their actions through a series of indicators and concrete measures, drawing in particular on the European Energy Award (EEA) label. Energy savings are identified through a detailed energy study, combined with an analysis of consumption data (bills, equipment, heating/air conditioning systems, etc.) and an assessment of energy performance indicators (EPIs). This method enables technical recommendations for improvement to be formulated, such as equipment renovation or system automation. However, its application in small and medium municipalities (SMMs) faces many challenges: the complexity of data integration, the need for advanced analysis tools, regulatory requirements such as ISO 50001, high deployment costs, and a lack of internal skills to effectively operate the system. These constraints hinder the effectiveness of Cit’ergie in SMMs, requiring simpler, faster approaches that are adapted to their limited resources [37,38,39,40].

To identify energy-saving opportunities with a Cit’ergie system, several key steps are involved (Figure 8).

2.6. Sequential Approach to Existing Methods

The identification of energy-saving potential scope (ESPS) by these systems is based on several methods aimed at selecting the energy-intensive units most relevant to energy optimization. One of the most common approaches is a sequential selection, focusing on units with the highest energy consumption. However, this method has its limitations, as it risks excluding certain low-consumption units which, due to their age or regulatory non-compliance, could nevertheless offer significant potential for energy savings.

To refine this selection, some methods incorporate additional criteria such as the state of obsolescence and regulatory compliance of equipment. A data sheet is generally drawn up to evaluate these criteria, with a rating used to classify energy-hungry units. This assessment then results in a multi-criteria classification, combining energy consumption, obsolescence and regulatory compliance. However, integrating these criteria requires additional resources and often time-consuming engineering activities.

Based on this classification, a weighted ranking of energy-intensive units is carried out using tools such as Excel to identify the most relevant energy-saving opportunities. Depending on the resources available, municipalities can then select the first ‘n’ opportunities to create an optimized ESPS. The advantage of this approach lies in its ability to include certain units that were initially underestimated due to their low consumption but which present a high potential for optimization thanks to the criteria of obsolescence and regulation.

Thus, the identification of an ESPS cannot be limited to a single criterion of energy consumption. A combined approach is essential to ensure a more relevant and effective selection.

2.7. Limits of Existing Systems Methods

Metering-based methods for identifying energy-saving potential scope (ESPS) have a number of intrinsic limitations that can hamper their effectiveness. Firstly, traditional metering systems are often based on technologies that do not allow for sufficiently fine-grained and detailed data collection. This can lead to excessive aggregation of data, making it difficult to accurately identify sources of energy waste. When considering a large number of community assets, this limitation is exacerbated by the growing complexity of data management and analysis, as well as, in particular, the identification of energy-saving potential scope (ESPS). Figure 9 shows a simulation based on energy consumption data for the town of Quesnoy Sur Deûle (QSD).

The graph in Figure 9 shows the evolution of the number of energy-saving potential scopes (ESPSs) as a function of the number of energy-consuming units in the municipality of Quesnoy-Sur-Deûle (QSD). It can be seen that

• For one energy-intensive unit, no ESPS has been identified.
• For five energy-intensive units, 30 ESPSs can be implemented.
• From upwards of eight energy-intensive units, the number of ESPSs explodes: 254 ESPS for eight units, 510 ESPS for nine units, and 1022 ESPS for ten units.

The number of ESPSs is growing exponentially. Indeed, as long as the number of energy-intensive units is low (between 1 and 5), the number of ESPSs remains limited, but as soon as a total of 6 to 10 energy-intensive units is reached, the realization of ESPS becomes very important and complex. This reveals that energy interactions between units increase drastically, making it difficult to identify optimal scope accurately by hand. This analysis highlights the limitations of manual methods for identifying ESPS once a certain threshold of energy-intensive units is reached. The exponential increase in the number of possible combinations calls for the use of advanced analysis tools to optimize community energy management quickly and efficiently.

To support our finding that existing systems are manual, our field survey revealed the solutions and software currently used by municipalities to identify the ESPS (Figure 10).

The results of this survey reveal a heavy reliance on manual methods, particularly spreadsheets (77%), and a virtual absence of advanced software solutions for the identification of ESPS. Only a minority (31%) combine a metering system with a spreadsheet, and no respondents use energy information systems (EISs) or software solutions to improve the efficiency of this task. These results underline the crucial need to modernize the tools used by municipalities to optimize their energy-saving efforts.

The need to assess numerous buildings and infrastructures simultaneously means a significant increase in the time and human and material resources required. Teams need to be sufficiently trained and available to install, monitor, and maintain metering devices, which can represent a high cost for municipalities, especially smaller ones. In addition, traditional metering systems can suffer from accuracy and reliability problems. Sensors and meters can be subject to measurement errors, failures, or drift over time, compromising the quality of the data collected. These problems can lead to inaccurate assessments of potential energy savings, limiting the effectiveness of any corrective action taken. Finally, interpreting the data from these systems often requires advanced technical skills, which can represent an additional challenge for municipalities which do not have the resources to hire or train qualified personnel. In short, although metering systems are essential for the identification of ESPS, their intrinsic limitations call for technological and organizational improvements to maximize their effectiveness.

3. Methodology of the New Solution

Our approach is based on a combination of empirical observations, mathematical concepts, and models of set theory with accessible statistical criteria, eliminating the dependence on the time-consuming processes of traditional approaches and field surveys, integrating a critical realist epistemological posture.

This approach enabled us to expose the shortcomings of existing systems, in this case the local measurement system (LMS), Cit’ergie (a management and labeling program for municipalities) and the Real Estate and Energy Master Plan (REMP) system, used in energy diagnostics for municipality. With this in mind, we followed an inductive approach, starting with the observation of the perceived reality of our case study. From this observation and an empirical generalization, as well as field surveys, we developed a model based on mathematical science theory to improve existing systems.

These inductive approaches (Figure 11) [41] and abductive (Figure 12) approaches were necessary, as the development of the new solution’s methodology is part of an emerging field of research, where academic references remain limited and theory concerning the inadequacy of existing systems is poorly developed or even non-existent in the scientific academic literature. In the absence of specific methodologies in the literature, this study adopted an innovative combinatorial approach, relying on set theory to generate and analyze energy-intensive sub-assemblies. In addition, the simplified criteria of average consumption and upper quartile, introduced in this research, constitute an original and accessible alternative to traditional approaches, which are often time-consuming and complex.

3.1. A Combinatorial Approach to the New Solution Method

Identifying energy-saving potential scope (ESPS) is a major challenge for municipalities wishing to optimize their energy consumption through energy management systems that comply with current standards, such as ISO 50001. However, traditional approaches, generally based on criteria such as energy consumption, obsolescence, and regulatory compliance of energy-intensive units, require significant resources and time-consuming engineering activities, making them particularly complex to implement for the energy management of small and medium municipalities (SMMs) compliant with current standards, such as ISO 50001.

To overcome these limitations, the new solution method adopts a combinatorial approach enabling the rapid identification of an optimal ESPS while reducing the analytical workload. In contrast to the sequential approach, the new solution method introduces, within the ESPS, substitute criteria that are highly correlated to the classic criteria (obsolescence and regulatory compliance), as well as to energy consumption, in order to simplify the process of identifying the optimal ESPS. It is important to point out that data collection, while essential to any energy analysis, is a separate process and not part of the new solution methodology illustrated in Figure 13.

3.1.1. ESPS Constitution (Candidate Subsets)

The first step is to generate all possible subsets (scopes) of a set of energy-consuming units in a local authority, without including an empty set or a complete set. These scopes are made up of buildings or equipment, for example, with measurable energy consumption. The aim is to create several combinations of scopes that can represent an ESPS.

Let $E=\left\{e_1,e_2, e_3,\dots ..e_n\right\}$, all the energy-consuming units identified within a municipality.

The total number of possible subsets of $E$, including the empty set $\varnothing$ and the complete set $E$, is given by

$$\mathrm{T}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l} \mathrm{n}\mathrm{o}\mathrm{m}\mathrm{b}\mathrm{e}\mathrm{r}=2^n.$$

(1)

However, for the identification of ESPS, it is necessary to exclude

The empty set $\varnothing$, which contains no energy-consuming units.

The complete set $E$, which represents all the energy-consuming units and does not allow for an optimal ESPS.

The number of candidate sub-assemblies selected by the new solution is

$$\mathrm{N}\mathrm{o}\mathrm{m}\mathrm{b}\mathrm{e}\mathrm{r} \mathrm{o}\mathrm{f} \mathrm{s}\mathrm{u}\mathrm{b}\mathrm{s}\mathrm{e}\mathrm{t}\mathrm{s}=2^n-2.$$

(2)

To form these subsets, the new solution generates all possible combinations of elements from $E$ with size $k$, such that

$$1 \le k \le n-1$$

(3)

The number of subsets of size $k$ is given by the binomial coefficient

$$C\left(n,k\right)=\left({\displaystyle \genfrac{}{}{0pt}{}{n}{k}}\right)={\displaystyle \frac{n!}{k!\left(n-k\right)!}}$$

(4)

The set of candidate subsets (candidate ESPS) is therefore defined by

$$P\left(E\right)=\bigcup _{k=1}^{n-1}\left\{S_i\subseteq E :\left| S_i\right|=k\right\}$$

(5)

where

• $S_i\in P\left(E\right)$ is a subset of energy-consuming units from $E$.
• $P\left(E\right)$ is the set of eligible ESPS candidate subsets excluding the empty set $\varnothing$ and the complete set $E$.

3.1.2. Selection of Eligible ESPS (Filtering of Eligible Subsets)

An initial filter is applied to identify the scope that meets the threshold of 65% [42] of the organization’s total energy consumption, in line with the recommendations of the ISO 50001 standard, which is based on experience. Although this threshold is a benchmark, it is not rigid: municipalities can choose a higher threshold depending on their energy policy and energy management maturity.

Scopes meeting this criterion are considered “eligible scopes”, i.e., potential candidates for identification of the optimum ESPS.

• Calculation of energy consumption of sub-assemblies:

For each subset $S_i \subseteq E$, the total energy consumption $C\left(S_i\right)$ is calculated:

$$C\left(S_i\right)=\sum _{e \in S_i}C\left(e\right)$$

(6)

where $C\left(e\right)$ is the individual consumption of the energy-consuming unit $e$.

• Filtering eligible subsets.

$C\left(S_i\right)$ is the total energy consumption of subset $S_i$. Subset $S_i$ is considered an eligible ESPS if

$${\displaystyle \frac{C\left(S_i\right)}{C\left(E\right)}} \ge 0.65$$

(7)

where $C\left(E\right)$ is the total energy consumption of the complete assembly $E$.

Subsets that meet this criterion are then retained for the next stage of the new solution methodology.

3.1.3. Definition of Substitute Criteria

Rather than directly assessing obsolescence and regulatory compliance—criteria that require resources and in-depth technical studies—the new solution proposes an alternative approach (Figure 14).

• Calculated indicators are extracted from eligible perimeter variables and tested to determine their correlation with conventional criteria.
• Statistical analysis is performed using Pearson correlation coefficients and p-values, allowing for the identification of indicators most strongly associated with obsolescence and regulatory compliance.
• These indicators thus become the new criteria, called substitute criteria, for refining the ESPS selection.

Once the eligible ESPS have been identified, based on the threshold of 65% of total energy consumption, the next step is to calculate several key indicators that will serve as the basis for identifying substitute criteria. These criteria are designed to effectively replace the often time-consuming traditional variables of obsolescence and regulatory compliance.

This calculation phase was carried out during tests carried out in several municipalities, including Quesnoy-sur-Deûle, where 6623 ESPSs eligible for the certification scope of EMS (energy management system) were identified. For each eligible ESPS, the following variables were calculated and integrated into a table comprising seven main columns:

• Total ESPS surface area: Sum of the surface areas of the energy-intensive units making up the ESPS.
• Total ESPS energy consumption: Sum of annual energy consumption of all ESPS units.
• ESPS average energy consumption: Arithmetic mean of the energy consumption of the units making up the ESPS.
• Energy consumption per m² of ESPS: Ratio between the total energy consumption and total surface area of the energy-consuming units making up the ESPS.
• Quartile 75 of ESPS energy consumption: Value located at the 75th percentile of individual energy consumption of ESPS units, highlighting energy-hungry units with the highest consumption.
• ESPS total obsolescence value: Sum of the obsolescence indices of the units making up the ESPS.
• Total ESPS regulatory status value: Sum of ESPS unit regulatory compliance indices.

The data obtained are grouped together in a comprehensive table, which serves as the basis for calculating Pearson correlation coefficients and p-values (

Table 5. Correlation coefficients of ESPS variables to identify substitute criteria.

ESPS Variable Pairs	Pearson Coefficients	p-Value	Comments
Total surface area and total obsolescence value.	0.73	0	Promotes energy consumption.
Total surface area and total value of regulatory status.	0.76	0
Energy consumption per m 2 and total obsolescence value.	−0.54	0	Favors energy consumption, obsolescence, and regulation. Less effective criterion for energy-hungry ESPS units.
Energy consumption per m 2 and total regulatory value.	−0.64	0
Average energy consumption and total obsolescence value.	−0.81	0	Favors energy consumption, obsolescence, and regulation. Criterion effective for up to 5 energy-hungry units in an ESPS.
Average energy consumption and total regulatory value.	−0.65	0
Quartile 75 energy consumption and total obsolescence value.	−0.67	0	Favors energy consumption, obsolescence, and regulation. Criterion effective for up to 5 energy-hungry units in an ESPS.
Quartile 75 energy consumption and total regulatory value.	−0.55	0

). These statistical analyses enable us to identify the substitute criteria that correlate most strongly with the obsolescence and regulatory compliance variables.

The next step is a second filter applied to eligible ESPS, based on the identified surrogate criteria. The optimal ESPS is determined by maximizing the value of the chosen substitution criterion. This approach effectively selects the perimeter with the best potential for energy optimization.

The coefficients obtained reveal a negative, moderately strong, and statistically significant correlation (p-value < 0.05) between the variables analyzed, although correlation does not necessarily imply a causal relationship. We observe that older or dilapidated buildings seem, counter-intuitively, to consume less on a heritage scale, which may be explained by reduced usage. Furthermore, technical “age” and compliance with regulations do not systematically translate into higher energy consumption. These observations suggest that using average consumption and the 75th percentile as substitute criteria for obsolescence and regulatory status would be a relevant choice, in the fourth stage, for identifying the certification scope (CS) of the energy management system (EMS) according to ISO 50001.

In the rest of our study, these two criteria (average consumption and 75th quartile) will be used in the new solution method to help identify the optimum ESPS, taking into account energy consumption, equipment obsolescence, and regulatory compliance.

In the new solution method, average energy consumption and quartile 75 (Q3) are used as proxy criteria to simplify the identification of relevant ESPS.

Calculation of substitute criteria: For each eligible ESPS, two simplified criteria are calculated:

• Average consumption:

$$\overline{C}\left(S_i\right)={\displaystyle \frac{1}{| S_i|}} \sum _{e \in S_i}C\left(e\right)$$

(8)

• Upper quartile:

The upper quartile $Q_3\left(S_i\right)$ is determined by ordering the $C\left(e\right)$ of $S_i$ in ascending order, then selecting the value at the position corresponding to 75% of the workforce. Its determination follows the formula

$$Q_3=X\left({\displaystyle \frac{3\left(n+1\right)}{4}}\right)$$

(9)

where

✓

$n$ is the total number of data items.

✓

$X\left(i\right)$ denotes the value at position $i$ in the sorted series.

For data sets where the calculated position $i$ is not an integer, linear interpolation is performed to determine the exact value of $Q_3$. For example, if the calculated position is $i=8.25$, then

$$Q_3=X\left(8\right)+0.25\times \left(X\left(9\right)-X\left(8\right)\right)$$

(10)

The use of the 75 quartile thus makes it possible to prioritize energy-intensive units which, although less consumptive in appearance, may be critical due to their obsolescence or regulatory non-compliance. This ensures a more inclusive and robust approach to identifying relevant ESPS.

3.1.4. Selecting the Right ESPS

Among the eligible subsets, the optimal perimeter is the one that maximizes the statistical criteria defined above (mathematical formula of the 2 criteria in a contracted calculation):

$$S_i^{*}=arg \underset{S_i \in \mathit{eligible}}{\mathit{max}}\left\{\overline{C}\left(S_i\right),Q_3\left(S_i\right)\right\}$$

(11)

This scope $S_i^{*}$ has been selected as the optimum ESPS.

In other words, the relevant ESPS is selected by maximizing one of the two alternative criteria chosen by the energy decision-makers (mathematical formula of the 2 criteria in 2 separate calculations):

• Selection by maximum average consumption:

$$S_i^{*}=arg \underset{S_i \in \mathit{eligible}}{\mathit{max}}\left\{\overline{C}\left(S_i\right)\right\}$$

(12)

This criterion makes it possible to prioritize sub-assemblies (ESPS) whose energy-hungry units have a high average consumption, often associated with outdated or non-compliant infrastructures.

This scope $S_i^{*}$ has been selected as the optimum ESPS.

• Selection by maximum 75 quartile:

$$S_i^{*}=arg \underset{S_i \in \mathit{eligible}}{\mathit{max}}\left\{Q_3\left(S_i\right)\right\}$$

(13)

This criterion favors sub-assemblies (ESPS) containing energy-intensive units with particularly high consumption, even if these units are in the minority in the ESPS identified.

This scope $S_i^{*}$ has been selected as the optimum ESPS.

• Further explanations:

  ✓

  The maximum average consumption criterion identifies the ESPSs that, on average, consume the most energy, which is often linked to outdated or non-compliant equipment.

  ✓

  The maximum 75 quartile criterion focuses on energy-intensive units with the highest consumption, even if they are few in number, which is particularly useful when some units are under-utilized due to obsolescence or non-compliance.

  ✓

  The choice between these two criteria is left to the energy decision-maker, depending on local priorities and the specific objectives of the study or the community’s energy policy strategy.

3.1.5. Evaluation and Validation of the Relevant ESPS

The final step is to assess the relevance of the ESPS identified by the new solution by comparing it with a reference scope derived from an existing method within a community or organization.

• Calculation of the match rate between the relevant ESPS and the reference ESPS:

  $$\mathrm{C}\mathrm{o}\mathrm{r}\mathrm{r}\mathrm{e}\mathrm{s}\mathrm{p}\mathrm{o}\mathrm{n}\mathrm{d}\mathrm{e}\mathrm{n}\mathrm{c}\mathrm{e} \mathrm{r}\mathrm{a}\mathrm{t}\mathrm{e} = {\displaystyle \frac{\mathrm{N}\mathrm{u}\mathrm{m}\mathrm{b}\mathrm{e}\mathrm{r} \mathrm{o}\mathrm{f} \mathrm{u}\mathrm{n}\mathrm{i}\mathrm{t}\mathrm{s} \mathrm{c}\mathrm{o}\mathrm{m}\mathrm{m}\mathrm{o}\mathrm{n} \mathrm{t}\mathrm{o} \mathrm{t}\mathrm{h}\mathrm{e} \mathrm{r}\mathrm{e}\mathrm{l}\mathrm{e}\mathrm{v}\mathrm{a}\mathrm{n}\mathrm{t} \mathrm{E}\mathrm{S}\mathrm{P}\mathrm{S} \mathrm{a}\mathrm{n}\mathrm{d} \mathrm{r}\mathrm{e}\mathrm{f}\mathrm{e}\mathrm{r}\mathrm{e}\mathrm{n}\mathrm{c}\mathrm{e} \mathrm{E}\mathrm{S}\mathrm{P}\mathrm{S}}{\mathrm{T}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l} \mathrm{n}\mathrm{u}\mathrm{m}\mathrm{b}\mathrm{e}\mathrm{r} \mathrm{o}\mathrm{f} \mathrm{u}\mathrm{n}\mathrm{i}\mathrm{t}\mathrm{s} \mathrm{o}\mathrm{f} \mathrm{t}\mathrm{h}\mathrm{e} \mathrm{r}\mathrm{e}\mathrm{l}\mathrm{e}\mathrm{v}\mathrm{a}\mathrm{n}\mathrm{t} \mathrm{E}\mathrm{S}\mathrm{P}\mathrm{S}}}\times 100$$
• Evaluation of the correspondence rate according to the majority principle: If a strong intersection is observed between the scopes identified by the new solution and those of traditional methods, then the method is deemed effective and usable. Thus, the correspondence rate, a key evaluation indicator, will be deemed sufficient if the ratio of the number of common units of the relevant ESPS and reference ESPS to the total number of units of the relevant ESPS reaches a rate greater than or equal to 50% according to the traditional methods scenario; this will enable us to verify the accuracy and effectiveness of the new solution method.
• Validation and final decision: Energy decision-makers evaluate the identified scope, adjusting it, if necessary, in line with the local authority’s objectives and strategic policy. Once validated after evaluation, the relevant ESPS becomes the certification scope (CS) for the ISO 50001-compliant energy management system.

3.1.6. Schematic Summary of the Methodology of the Combinatorial Approach to the Solution

Figure 15 schematically summarizes a methodological approach to the ESPS identification process using the new solution.

3.2. Justifying the Combinatorial Approach

The use of this combinatorial approach ensures that no potential subset is overlooked, guaranteeing that all possible combinations of energy-intensive units are evaluated before applying the selection criterion. In addition, it allows us to identify energy-intensive units which, despite low individual consumption, may be relevant due to their obsolescence or regulatory non-compliance, thanks to the introduction of the average substitute criterion, as well as the 75 quartile criterion, which favors subsets containing energy-intensive units with particularly high consumption.

4. PG2E: A New Digital Solution for Identifying the Right ESPS

4.1. Introducing the PG2E Application

PG2E is an innovative, fully digital solution that uses a combinatorial approach to automatically identify the ISO 50001 certification scope based on monthly or annual energy consumption data and the definition of substitute criteria (Figure 16).

4.2. Automating the PG2E Process

The PG2E methodology is fully automated using a combinatorial programming algorithm implemented in Python 3. The algorithm follows the following steps:

• Generate all possible subsets using Python’s combinatorial library (itertools.combinations) (1) + (2) + (3) + (4) + (5).
• Calculation of $C\left(S_i\right)$ energy consumption for each subset and filtering of eligible ESPS: (6) + (7).
• Identification of substitute criteria (mean and quartile 75) for obsolescence and regulatory compliance using Pearson coefficients and p-values. Calculation of substitute criteria $\overline{C}\left(S_i\right)$ and $Q_3\left(S_i\right)$ for eligible subsets: (8) + (9) + (10).
• Automatic selection of the relevant perimeter (11) or ((12) and (13)).
• Evaluation and validation of the relevant ESPS.

The originality of this research lies in the development of PG2E, a new method and innovative digital solution based on set theory and simplified statistical criteria. It automatically identifies energy-saving potential scope (ESPS) from a single data source, notably energy consumption, with results comparable to conventional methods, while being faster, less costly, and better suited to small and medium municipalities (SMMs).

4.3. Case Study: Municipality of “Quesnoy-Sur-Deûle” (QSD)

4.3.1. Case Study Background

Quesnoy-sur-Deûle, a French municipality of 7000 inhabitants, is an ideal setting for testing the innovative PG2E solution. This community represents a common typology of small and medium municipalities (SMMs), characterized by

• A diverse range of buildings, including administrative, school, and sports facilities.
• Limited technical and financial resources for energy management.
• A regulatory obligation to comply with standards such as the tertiary sector decree and ISO 50001.

The case study conducted on the municipality of Quesnoy-sur-Deûle aims to assess the relevance and reliability of the PG2E method in a real-life context, by comparing its results with those of traditional approaches used by the local authority. In particular, the application of the REMP (Real Estate and Energy Master Plan) method to identify the reference ESPS revealed several issues:

• Complexity: Taking into account outdated infrastructures and regulatory compliance considerably slowed down the process of identifying ESPS.
• Excessive duration: The energy-saving potential scope project took nine months to complete, illustrating the slowness of conventional approaches such as REMP.

To verify the effectiveness of PG2E, this study, carried out in the municipality of Quesnoy-sur-Deûle, was selected on the basis of several criteria:

• Accessibility of the energy data needed to apply the PG2E method.
• Diversity of energy-consuming units, enabling a representative analysis of the municipality’s buildings.
• The local authority’s interest in and commitment to improving its energy efficiency and ISO 50001.certification.

4.3.2. Data Preparation and Preliminary Analysis

Before applying the PG2E method, a preliminary data analysis is required to better characterize the building stock in the municipality of Quesnoy-sur-Deûle. The data studied include

• Annual energy consumption in kWh of electricity (reference year for electricity consumption is 2021).
• Total surface area of buildings (in m²).
• The state of disrepair of the infrastructure, assessed on a scale of 0 to 100. A score above 50 indicates significant disrepair, while a score above 70 suggests that intervention or replacement should be considered within 2 to 3 years.
• The regulatory compliance of buildings is rated on a scale of 0 to 1. The closer the rating is to 1, the higher the level of non-compliance with regulatory requirements.

This preliminary analysis is essential for identifying the heterogeneities that influence energy consumption and optimization potential. By highlighting buildings requiring priority action, it improves the relevance of the criteria used by PG2E, notably average energy consumption and quartile 75.

The main aim of this stage is to provide an in-depth understanding of trends and disparities within the building stock in order to better interpret the results obtained by the PG2E method. By validating methodological choices, this analysis ensures that PG2E’s combinatorial approach is adapted to the specific features of the buildings studied.

Table 6. Data on electricity consumption in kWh, obsolescence, and regulatory status of buildings for 2021.

List of Study Elements: Electricity Consumption 2021 in kWh
Building Code	Surface M2	Old	Regulatory Status	Annual Consumption
CD02	601	75.00	0.14	13,279
CD03	207	76.25	0.00	10,278
CD04	700	57.04	0.43	16,363
CD05	670	62.96	0.19	8689
CD06	1070	53.42	0.29	7513
CD07	620	67.31	0.14	22,417
CD08	798	52.50	0.43	19,615
CD09	2275	56.30	0.29	85,228
CD11	1383	61.71	0.14	31,243
CD12	439	63.27	0.14	12,028
CD14	262	57.04	0.14	11,293
CD15	338	73.19	0.14	10,042
CD16	1004	57.96	0.14	69,680
CD17	1171	57.50	0.14	76,966
CD18	983	74.21	0.14	35,575

shows data on the annual energy consumption, obsolescence, and regulatory status of buildings in the municipality of Quesnoy-sur-Deûle. These data will serve as a basis for evaluating the effective application of the PG2E method and comparing the results obtained with those of traditional approaches.

Data on the obsolescence (rated out of 100) and regulatory status (rated out of 1) of buildings have exceptionally been integrated into the table structure for a correlation assessment of the alternative criteria of average consumption and quartile 75 of eligible ESPS, thus enabling us to identify the certification scope (CS) dedicated to the energy management system (EMS).

A preliminary analysis has been carried out on the annual consumption, surface area, obsolescence, and regulatory compliance of buildings in the municipality of QSD (Figure 17).

The graphs in Figure 17 show the distribution of normalized values for four variables relating to buildings in the town of Quesnoy Sur Deûle (QSD): annual energy consumption, total surface area, dilapidated condition, and regulatory status. Overall, we observe the following analysis:

• The graph shows that energy consumption (red curve) and building surface area (blue curve) have different distributions, with notable deviations, suggesting that energy consumption could be influenced by factors other than simple building size, such as obsolescence or regulatory status.
• Obsolescence (orange curve) and regulatory status (green curve) appear to play an important role in the variability of energy data, with obsolescence particularly asymmetrical, indicating that a proportion of buildings are significantly older.

In conclusion, the distributions presented in the graphs (Figure 17) reveal significant variability in the characteristics (obsolescence and regulatory compliance) of the buildings analyzed, highlighting the need for a differentiated approach to optimizing energy consumption management. Larger, older buildings could benefit from targeted interventions to improve their energy efficiency, hence the importance of quickly identifying them using a combinatorial approach based on a simple, effective methodology.

4.3.3. Results Obtained with REMP Scenarios

In our case study, the commune of Quesnoy-sur-Deûle (QSD) identified three reference ESPSs through their REMP system according to political mandates. These reference ESPSs, identified after 9 months of activity, are as follows:

• P_{05 Réf}. = {CD09, CD08, CD04, CD17, CD16}: A reference scope of five energy-consuming units.
• P_{08 Ref}. = {CD09, CD08, CD04, CD17, CD16, CD06, CD11, CD14}: A reference scope of eight energy-consuming units.
• P_{10 Réf}. = {CD09, CD08, CD04, CD17, CD16, CD06, CD11, CD14, CD07, CD18}: A reference scope of 10 energy-consuming units.

Our job is to identify the best ESPS, which could correspond to the different reference ESPSs, using the innovative PG2E solution of the new method.

4.3.4. Description of the Execution Scheme and Result of PG2E

As a reminder, the PG2E solution is based on a combinatorial methodology that enables an efficient identification of the certification scope (CS) for the energy management system (EMS) according to the ISO 50001 standard. As input, it takes the consumption data for energy-intensive units, their characteristics (total surface area, obsolescence, regulatory compliance), and several dynamic parameters: name of local authority, reference year, 65% minimum consumption threshold, and choice of substitute criteria (average consumption or 75 quartile). By following five methodological steps, PG2E produces an optimized certification scope (CS) with a summary report presenting the criteria used, the results obtained, and a list of relevant ESPSs for informed decision-making. Figure 18 illustrates this flow chart.

The application of the PG2E tool identifies 6623 potential ESPS candidates for the certification scope (CS). Using the average consumption criterion, PG2E identifies the CS among the eligible ESPS (

Table 7. Detailed CS results identified by the PG2E solution.

Certification Scope (CS) Details
Building Code	Annual Consumption (kWh)	Old	Condition Regulatory	Comments
CD09	85,228	56.30	0.29	Presence of aging. Potential for savings
CD11	31,243	61.71	0.14	Presence of aging. Potential for savings
CD16	69,680	57.96	0.14	Presence of aging. Potential for savings
CD17	76,966	57.50	0.14	Presence of aging. Potential for savings
CD18	35,575	74.21	0.14	High age and potential. Change in 2 to 3 years

):

P_conso-m = {CD09, CD11, CD16, CD17, CD18}.

Table 7 details the certification perimeter identified by the PG2E solution, which will then be compared with the municipality’s reference perimeters in the “Benchmarking results” section of the article.

These five buildings have the highest energy consumption of the fifteen units studied, which justifies their being seen as priorities in terms of optimization potential. All buildings are over 50 years old, which means they are obsolete and potentially a source of energy loss. CD18 (74.21) exceeds the critical threshold of 70, suggesting that it requires improvement within 2 to 3 years, indicating a high potential for savings. The poor regulatory status of CD09 (0.29) is another signal. The combination of high consumption and average to high obsolescence confirms that these buildings are not only energy-hungry, but also candidates for corrective action.

The PG2E solution does more than simply classify buildings according to their absolute energy consumption. It applies a combinatorial optimization logic based on proxy criteria—average consumption and upper quartile (Q3)—which have shown a strong negative correlation with obsolescence and regulatory status (values between −0.55 and −0.81 with p-values < 0.05). These correlations justify their use as indirect indicators of energy-saving potential.

Figure 19 provides a summary of the parameters used to obtain certain results, as well as the ISO 50001-compliant certification scope (CS), generated as a report by the PG2E solution. This report also includes, in Figure 20, a list of the top 10 relevant ESPSs, made available to decision-makers for the analysis and orientation of the energy policy of the municipality of Quesnoy-Sur-Deûle.

The scope of certification (CS) of the SME according to ISO 50001 remains identical whether the criterion of average consumption is used or that of the upper quartile, which is as follows:

Quartile 75 consumption P_qrt-75 = {CD09, CD11, CD16, CD17, CD18}.

5. Results of the Comparative Study of ESPSReference and ESPSPg2e

5.1. Scenario 10: The Top 10 Energy-Generating Units

With a reference perimeter of 10 energy-intensive units identified for the town of Quesnoy sur Deûle, we observed a 100% correspondence rate (

Table 8. Results of scenario 10 with the PG2E solution and the ESPS_Ref. P₁₀.

P10—Electricity 2021
Compared Element	QSD Reference ESPS (10 Units)	ESPS of the PG2E Solution Identified (5 Units)	Matching Rate	Bias
List of units:	CD09, CD08, CD04, CD17, CD16, CD06, CD11, CD14, CD07, CD18	CD09, CD11, CD16, CD17, CD18	100%	0%
Criterion used Average:	✓ All PG2E units present in the reference
Criterion used Quartile 75:	✓ All PG2E units present in the reference

) with the “PG2E” tool according to the criteria of average and upper quartile energy consumption for the scope identified by the solution.

5.2. Scenario 08: The Top Eight Energy-Generating Units

With a reference perimeter of eight energy-intensive units identified for the town of Quesnoy sur Deûle, we observed a correspondence rate of 83% (

Table 9. Results of scenario 08 with the “PG2E” solution and the ESPS_Réf. P₀₈.

P08—Electricity 2021
Compared Element	QSD Reference ESPS (8 Units)	ESPS of the PG2E Solution Identified (5 Units)	Matching Rate	Bias
List of units:	CD09, CD08, CD04, CD17, CD16, CD06, CD11, CD14	CD09, CD11, CD16, CD17, CD18	83%	17%
Criterion used Average:	✓ 4/5 PG2E units present in the reference
Criterion used Quartile 75:	✓ 4/5 PG2E units present in the reference

) with the “PG2E” tool according to the criteria of average and upper quartile energy consumption of the perimeter identified by the solution.

5.3. Scenario 05: The Top Five Energy-Generating Units

With a reference scope of five energy-intensive units identified for the town of Quesnoy sur Deûle, we observed a correspondence rate of 60% (

Table 10. Results of scenario 05 with the “PG2E” solution and the ESPS_Réf. P₀₅.

P05—Electricity 2021
Compared Element	QSD Reference ESPS (5 Units)	ESPS of the PG2E Solution Identified (5 Units)	Matching Rate	Bias
List of units:	CD09, CD08, CD04, CD17, CD16	CD09, CD11, CD16, CD17, CD18	60%	40%
Criterion used Average:	✓ 3/5 PG2E units present in the reference
Criterion used Quartile 75:	✓ 3/5 PG2E units present in the reference

) with the “PG2E” tool according to the criteria of average and upper quartile energy consumption of the perimeter identified by the solution.

It is important to emphasize that all the scenarios to identify the certification perimeters compliant with the ISO 50001 standard were carried out in just half a day, thanks to the use of the innovative PG2E solution. This rapid completion was made possible by the prior collection of data and the identification of the reference certification perimeters associated with each scenario.

5.4. Quantitative Comparison Between REMP Method and PG2E Solution for ESPS Identification

Table 11. Quantitative comparison between REMP method and PG2E solution to identify ESPS.

Comparison Criteria	Current Method (REMP)	Innovative Method (PG2E)	Estimated Gain
Average process duration	3 to 18 months	<1 day (execution < 1 min)	>95% time saving
Human resources mobilized	Multidisciplinary team of engineers	1 operator with basic energy skills	Significant reduction
Necessary tools	Excel, technical expertise, audits	PG2E digital tool + consumption file	Streamlining the process
Estimated cost of diagnosis for SMM	Between EUR 10,000 and EUR 25,000	< EUR 1000 (depending on the scope)	Up to 80% cost reduction
Correspondence rate with the reference ESPS	Reference used	60% to 100% depending on the scenarios	Equivalent accuracy
Process automation	No	Yes	Complete automation
Ability to generate multiple relevant ESPS	No	Yes (classification of ESPS by criteria)	Broad exploratory approach
Integrated criteria (consumption data only)	No (multiple data, manual diagnostics)	Yes (annual consumption, substitute criteria)	Ease of data exploitation

presents a detailed quantitative analysis based on several criteria: ESPS correspondence rate, process duration, estimated cost, resources mobilized, level of automation, etc.

This comparison highlights the concrete advantages of the PG2E solution in terms of time savings (>95%), cost reduction (up to 80%), and the ability to automatically identify several relevant ESPS while maintaining a satisfactory level of accuracy.

The results obtained from this comparative study confirm that the PG2E method offers accuracy comparable to traditional methods while significantly reducing the time and resources required.

6. Discussions

6.1. Observations and Analysis of the Case Study

In this case study, the innovative PG2E solution was applied to the commune of Quesnoy-sur-Deûle to assess its effectiveness in identifying energy-saving potential scope (ESPS). The methodology is based on a combinatorial approach to building ESPS, using a threshold of 65% of total ESPS consumption to select eligible ESPSs and then using average consumption and upper quartile as priority criteria to identify the relevant ESPS or certification scope (CS).

The results were analyzed across three scenarios, comparing the ESPS identified by PG2E with the reference ESPS established by the municipality. For a large scope (Scenario 10), the reference ESPS captured 100% of the ESPS units identified by the PG2E solution, thus validating the relevance of the criteria used. For an intermediate scope (Scenario 08), the match rate was 83%, demonstrating a high degree of compatibility between the units selected by PG2E and those of the reference ESPS. On the other hand, for a more restricted scope (Scenario 05), the match rate dropped to 60%, revealing the limits of the solution when the perimeter becomes more selective. These results show that PG2E is particularly effective in identifying ESPS with extended configurations. However, adjustments to the criteria could be considered to improve matching in smaller perimeters. The 65% threshold recommended by ISO 50001 proved to be relevant for large scope but could be modulated according to the diversity of infrastructures. What is more, the average consumption and upper quartile criteria proved robust enough to effectively prioritize energy-intensive scope, including those comprising low-consumption units that are potentially outdated or not compliant with regulations. PG2E therefore appears to be a suitable solution for municipalities wishing to quickly identify relevant ESPS while offering scope for improvement to adapt to more specific configurations.

6.2. Evaluation the Innovative PG2E Solution

Analysis of the PG2E solution highlights several key advantages, including its simplicity and accessibility, enabling rapid, automated identification of energy-saving potential scope (ESPS) without requiring significant human or financial resources. Unlike traditional methods based on complex technical audits, PG2E offers a streamlined approach tailored to small and medium municipalities (SMMs). One of its major advantages is the significant time saving, estimated at over 95% (Appendix A) compared with conventional methods, a budget saving of up to 80% (Appendix B), and the reduction in resources mobilized, with simulations taking just a few seconds compared with several months for conventional methods such as REMP (Appendix C), facilitating proactive decision-making and reducing the cost of energy diagnostics. What is more, its flexibility and adaptability mean that the perimeters identified can be adjusted according to local priorities, thanks to the introduction of dynamic thresholds, making PG2E scalable even for restricted scope.

In addition to standards such as ISO 50006, which provide a framework for measuring and monitoring energy performance using advanced indicators (EnPIs) with a view to continuous improvement [43], the PG2E solution comes into play upstream, facilitating the initial identification of a relevant scope for deploying an EMS. It thus constitutes a strategic preparatory tool, strengthening the foundations on which the normative requirements of energy management will then be applied.

However, certain limitations were identified, notably the solution’s dependence on data quality, as an incomplete or inconsistent database could affect the accuracy of results. Furthermore, while PG2E is well suited to SMMs, its application in larger communities with more complex infrastructures would require methodological adjustments to incorporate a more detailed analysis of interactions between infrastructures. Another point to note is the absence of exhaustive technical assessments, since PG2E does not take into account elements such as infrastructure aging, regulatory status, or maintenance needs, which may limit its scope in certain situations even if it uses substitute criteria.

Table 12. Comparative summary of existing systems and the PG2E solution.

Criteria	Existing Systems	PG2E Solution
Methodological approach	Unit-by-unit sequential approach based on detailed analysis of building and energy equipment characteristics to build up the reference ESPS.	Combinatorial approach based on a set of units (ESPS) meeting the threshold of 65% of total energy consumption and statistical criteria (mean and quartile 75) to identify the relevant ESPS.
Main objective	Identify the most energy-intensive buildings and analyze their performance in detail.	Quickly identify ISO 50001-compliant energy-saving potential scope (ESPS).
Accuracy of results	Very accurate, but depends on data availability and quality.	Results comparable with existing systems using relevant statistical criteria, but dependent on data availability and quality.
Analysis time	Long (requires in-depth data collection and analysis).	Reduced in the area of the relevant ESPS identification process, but dependent on data collection activities.
Resource requirements	Important (technical expertise, specialized software, energy audits).	Limited (no need for complex technical analyses, accessible to SMM).
Complexity of implementation	High (requires specific skills and advanced tools).	Low (simplified methodology, suitable for communities with few resources).
Accessibility for MMS	Difficult to implement due to cost and technical complexity.	Easy to adopt thanks to a simplified, fast approach.
Taking outliers into account	Based on in-depth analysis of buildings and their consumption history.	Identified by statistical criteria (mean and upper quartile), without requiring in-depth analysis.
Adaptability and flexibility	Requires frequent database updates and manual analysis.	Adaptable method, with the possibility of adjusting thresholds and criteria as required.
Difficulties in collecting data	YES. Lack of data available at municipal level with a semi-automated system.	YES. There is a lack of available data at municipal level, but it can integrate an energy information system (EIS) to automate and propose relevant ESPS.

summarizes a comparison between existing systems and the PG2E system.

7. Conclusions and Outlook

Small and medium municipalities (SMMs) face a number of obstacles in identifying their energy-saving potential, confronted as they are with conventional methods that are time-consuming (between 3 and 18 months), costly, and often ill-suited to their specific needs.

The work presented in this article highlights the limitations of current approaches for efficiently identifying energy-saving potential scope (ESPS). Faced with this observation, the PG2E solution is positioned as an innovative, rapid, and easily deployable digital alternative, designed to meet the constraints of SMM.

Tests carried out on data from the commune of Quesnoy-sur-Deûle confirm the performance of PG2E: success rates of 60% to 100% are achieved depending on the perimeters analyzed, with time savings of over 95% and an estimated reduction in diagnostic costs of up to 80%.

Unlike a simple prioritization of consumption, PG2E is based on an optimized combinatorial approach incorporating substitute criteria—average consumption and upper quartile (Q3)—which are strongly correlated with indicators of obsolescence and regulatory status (with negative coefficients between −0.55 and −0.81 and significant p-values). This enables it to identify not only energy-hungry units, but also those with high energy-saving potential.

Finally, the integration of additional variables and validation in other territories would consolidate PG2E’s effectiveness and extend its use to a wider range of communities committed to the energy transition.