Identification of District Heating and Cooling System Archetypes: a Novel Approach Applied to a Case Study in Switzerland

Identification of district heating and cooling system archetypes: a novel approach applied to a case study in Switzerland Luca Brauchli a,* , Núria Duran Adroher a , Willy Villasmil b , Markus Auer b , Edward Lucas a , Philipp Schuetz a , J¨ org Worlitschek a a Institute of Mechanical Engineering and Energy Technology, Lucerne University of Applied Sciences and Arts, Technikumstrasse 21, 6048, Horw, Switzerland b Institute of Building Technology and Energy, Lucerne University of Applied Sciences and Arts, Technikumstrasse 21, 6048, Horw, Switzerland ARTICLE INFO Handling editor: Henrik Lund Keywords: District heating and cooling District energy Decarbonisation Archetype Clustering Exergy demand ABSTRACT District Heating and Cooling (DHC) systems are central to the decarbonisation of thermal energy supply. For efficient development and assessment of decarbonisation strategies amongst these diverse systems, it is essential to identify representative system archetypes that can serve as proxies for broader types in targeted case studies. Existing work often focuses on single-criteria and supply-side classifications, while overlooking demand-side diversity and the role of exergy in evaluating system performance. This study introduces a novel multi-criteria methodology for the demand-based characterisation and grouping of DHC districts, with exergy as a central metric. The method proposes four criteria to characterise supply regions on GIS-based data: exergy demand density (in place of traditional energy demand), degree of connection to reflect densification potential, and two building stock indicators reflecting ownership patterns and usage types to incorporate a social dimension. The methodology applies a K-medoids algorithm to group similar systems and identify representative archetypes within each group. To demonstrate the approach, it was applied to publicly available data on existing Swiss DHC systems. As the required data on supply regions was not readily available, it was first generated through a GISbased analysis, using DBSCAN clustering to define district boundaries based on building-level information. This enabled the application of the proposed criteria and the identification of five distinct archetypes for Swiss DHC systems. Due to the slightly delayed perspective of the Swiss dataset, recent developments such as lowtemperature networks are underrepresented while capturing the more established, higher-temperature networks where decarbonisation needs remain particularly significant. 1. Introduction In the pursuit of enhancing energy efficiency and sustainability, understanding the dynamics and characteristics of district heating and cooling (DHC) systems has become crucial. DHC systems play a significant role in the efficient and fossil-free distribution and management of thermal energy. Despite their importance, comprehensive methodologies for systematically identifying and categorising these systems remain limited. A meaningful classification must include not only technical supply-side aspects but also non-technical - especially demand-side - characteristics, as demand forms the basis for evaluating supply structures. While extensive efforts have been made to identify and categorise buildings according to various attributes, similar systematic approaches for DHC systems are sparse. In this paper, the term DHC systems is used generically, although in many specific cases it refers to pure heating applications. DHC systems are generally described by a broad range of technical attributes. Books and guidelines like those proposed by Nussbaumer et al. [1], Rutz et al. [2] or Oppermann et al. [3] give a good overview on technical and economical properties of the network infrastructure. Hangartner et al. [4], Sulzer et al. [5] and Werner [6] clearly define a vocabulary that addresses relevant technical attributes. Frederiksen and Werner in Ref. [7] and Lund et al. in Ref. [8] introduced the concept of network generations, which has been broadly adopted by the scientific community. Generational classification represents the continuous reduction of network temperatures along with increasing efficiencies and shares of renewables, ranging from early high temperature steam networks (1st generation) to low and ultra-low temperature networks in later 4th and 5th generations. Network temperature is perhaps the most relevant attribute to distinguish between network generations. A similar * Corresponding author. E-mail address: [email protected] (L. Brauchli). Contents lists available at ScienceDirect Energy journal homepage: www.elsevier.com/locate/energy https://doi.org/10.1016/j.energy.2025.137159 Received 26 March 2025; Received in revised form 12 June 2025; Accepted 15 June 2025 Energy 333 (2025) 137159 Available online 26 June 2025 0360-5442/© 2025 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license ( http://creativecommons.org/licenses/by/4.0/ ). approach, more focused on low temperature networks (i.e. operating below 60 ◦C) was introduced by Hangartner et al. [4]. The proposed classification distinguishes networks based on four temperature levels, which determine their applicability for space heating, cooling, and domestic hot water production. Apart from network temperature, networks are often distinguished by size as proposed by Rutz et al. [2] or by energy source as suggested by Persson and Werner [9]. Triebs et al. [10] developed a data-based multi-criteria classification on CHP-fired networks in Germany. Data of 140 DHC systems was collected and an agglomerative clustering with 4 dimensionless key figures, focusing on different technical characteristics of the CHP units, was used to identify 8 types of systems. Werner [6] analysed an inventory of 165 implemented, planned, or proposed cases of 4th generation networks and buildings from 19 countries, resulting in 6 main groups of system configurations. Those classifications and configurations are focused on technical supply properties of DHC systems. Nevertheless, demand properties are essential to define pathways for the decarbonisation of districts. Exergy will play a central role in the characterisation of demand, since it offers a more accurate measure to assess the relevant efficiency (exergetic instead of energetic) of DHC systems and reveal potentials not visible by energy-based views alone, as correctly concluded by Kotlanien et al. [11]. This metric has increasingly been used to assess the efficiency of DHC systems, particularly in case studies such as those by Li et al. [12] and Veyron et al. [13], but also in more general works like Gong & Werner in Ref. [14] or Topal et al. in Ref. [15]. Given the heterogeneity of demand structures, a reduction in complexity is needed to generalise insights. GIS-based approaches are widely used in DHC research for automated layout design and optimisation, as shown by Tzouganakis et al. [16] and Chicherin [17]. Vrain et al. [18] apply spatially explicit methods for scenario development. A key input for such studies is the location and characteristics of existing DHC systems, as compiled by Pelda et al. [19] in a district heating atlas for Germany. However, these datasets mainly focus on technical aspects such as energy use, pipe length, and temperatures, often overlooking demand-side features. On a more high-level perspective, Mun´ can et al. [20] contributed a literature-based assessment that, beyond technical aspects, also considers economic and regulatory frameworks across European countries. In building energy modelling (BEM), the use of building archetypes is a well-established strategy for reducing complexity. The study of building archetypes aims to identify representative models that capture the common characteristics of buildings within a specific context using various approaches [21]. This context could be defined by factors like building type, climate zone, construction period or energy performance and the approaches strongly depend on the availability of information and the field of interest. The systematic identification of archetypes within BEM using clustering techniques has proven effective in simplifying complex energy systems, enabling targeted interventions [22] and allowing comparison of results using the same basis. For instance, Alrasheed & Moushed [23] emphasises the role of archetype classification using the k-prototype clustering method to improve energy performance simulations. Similarly, Dahlstr¨ om et al. [24] discuss the benefits of a novel multi-parameter cluster analysis approach for identifying representative building archetypes within the Swedish residential building stock. Inspired by BEM methodologies, this study introduces a novel, archetype-based approach to identify and characterise representative DHC systems. Rather than attempting to optimise individual DHC systems in isolation, the archetype approach enables the identification of a manageable set of representative systems that approximate the diversity of system configurations in the most effective way. The methodology presented is applied to the case study of Switzerland to identify a suitable number of Swiss DHC system archetypes. Importantly, the methodology itself is region-independent and can be applied to any regional/ supra regional context. This study does not provide system-level optimisation or regulatory guidance. Instead, it enables identification of DHC archetypes through a comprehensive, demand-oriented, and not purely technical lens. These archetypes offer a foundation for technical, economic, and regulatory analysis, allowing findings to be projected onto the wider heating and cooling sector. This enhances understanding and supports more effective design, planning, and policymaking. 2. Methodology DHC systems exhibit a range of characteristics that can be analysed from technical, economic, and social perspectives. This study distinguishes systems based on two property groups: supply infrastructure and district characteristics. With the perspective of these characteristics, archetypal networks and archetypal districts can be identified, which in combination define an archetype of DHC systems as illustrated in Fig. 1. Supply-side properties include generation capacity, energy sources, and network infrastructure – typically tied to technical and economic factors. Demand characteristics however, concern consumption behaviour, exergy requirements and the profiles of different building types within the district, which also reflect social characteristics. Most studies focus on supply-side aspects or individual networks, often overlooking comparative or socio-economic insights. Thus, our research pivots towards a demand-side perspective, recognising that supply systems exist to meet demand efficiently. Consequently, the structure and characteristics of the energy demand are crucial for a meaningful comparison of DHC systems. By examining the demand-side, we acknowledge that the efficiency of the network should be measured by its ability to meet the energy requirements of the connected buildings effectively and sustainably, and that the successful implementation of DHC systems depends on more than just technical aspects. The novelty of this work lies not in any single new metric, but in the integration of multiple underused perspectives into one framework. •A demand-oriented lens that includes technical and social dimensions (e.g. building type and use as proxies for ownership). •Use of exergy demand density instead of traditional energy demand density. While these aspects exist individually in literature, their combined application in a comparative, multi-district framework represents a new methodological contribution. Exergy analysis is common in HVAC but applying exergy demand density to define DHC supply areas remains rare. 2.1. Description of districts A method is proposed that identifies the buildings connected to the Fig. 1. Basic distinction of DHC system characteristics into demand and supply side. From their perspective, archetypal districts and archetypal networks can be identified, which interact with each other and define the archetypes of DHC systems as combinations of the two sides. L. Brauchli et al. Energy 333 (2025) 137159 2 DHC networks, delineates the served areas, and describes districts based on building characteristics. First, district area must be spatially defined. Knowing the buildings connected to a DHC network, the area is delineated by drawing a polygon around the building cluster using a defined distance and method. We propose using contiguous 100 m × 100 m geo raster cells, as shown in Fig. 2. This is inspired by the common use of hectare grid cells in Switzerland for area-based analyses, such as waste heat potential and heat demand mapping by Chambers et al. [25] and Schneider et al. [26]. Cells are centred on the cluster, and only those containing at least one connected building are included. Within this polygon, both connected and non-connected buildings are considered. For each building, the exergy demands are estimated based on location, building type, age and size. Further insights are provided in chapter 2.2. Based on this demand data and additional building information, four key performance indicators (KPI) are derived for each cluster: (1) The degree of connection (DOC) weighted by exergy demand, quantifying the share of demand that is currently supplied by the network and indicates the potential for enhancing network connectivity within the area; (2) exergy demand density, defined as the total exergy demand per unit area, a novel metric proposed to replace the traditional energy density indicator thereby offering a more precise measure to incorporate resource efficiency considerations; (3) share of residential energy reference area (ERA), indicating variation in user profiles; (4) share of single-family houses amongst residential buildings, offering insights into ownership patterns. Exergy demand constitutes a central part of KPIs (1) and (2) and the estimation methodology is described below. 2.2. Exergy demand estimation As stated by Gong & Werner in Ref. [27], exergetic efficiency was less relevant during the era of fossil-fuelled heating systems. For future systems relying on renewables and waste heat, using high-quality sources (e.g. gas, biomass) to produce low-quality heat for building application is inefficient. Thus, evaluating networks by their exergetic performance is essential for cost-effective energy transitions. While the exergy approach remains rarer in literature than energetic approaches, some literature deals extensively with an exergetic view on thermal Fig. 2. Example of 100 m ×100 m raster field to capture the supply area of a network as a polygon. Red dots indicate buildings connected to the network; yellow dots the non-connected buildings within the polygon; grey dots the nonconnected buildings outside the polygon. Fig. 3. Heat transfer station with the flow and return temperatures on both the primary and secondary side of the system. Table 1 Definition of flow and return temperatures depending on building age and assumed heat transmission system. Old radiators require the highest flow temperatures. Floor heating has a lower temperature difference between flow and return. Construction Period Heat transmission system Flow temperature Return temperature ≤1970 Radiator 70 ◦C 343 K 55 ◦C 328 K 1971–1990 Radiator 60 ◦C 333 K 45 ◦C 318 K 1991–2005 Radiator 50 ◦C 323 K 35 ◦C 308 K ≥2005 Floor heating 35 ◦C 308 K 25 ◦C 298 K Fig. 4. Arbitrary example of a point cloud in a 2-dimensional space and three groups/medoids to be identified. Colour indicates affiliation to group A, B, or C, and the filled circles mark the points that form the medoid for each group. The algorithm minimizes the sum of all distances between all points within each group and their corresponding medoids. L. Brauchli et al. Energy 333 (2025) 137159 3 networks. Gong & Werner [27] calculate exergy demand E by heat demand Q multiplied by an exergy factor ε : E=Q⋅ ε If unknown, heat demand must be estimated based on building properties and location. The basic principle consists of estimating the annual area specific space heating and hot water demand of a building based on type, age and location. This value must then be multiplied by given or estimated building energy reference area (ERA). For Swiss residential and service building stock, ERA is evaluated using the Fig. 5. Network count by main energy source [36]. Biomass includes logs, wood chips, pellets, and biogas. Location-bound low-temperature sources include lake and river water, groundwater, wastewater, and shallow geothermal. Fossil sources include heating oil, natural gas, and natural gas cogeneration plant. Other waste heat includes waste heat from nuclear power plant and tunnel waste heat. Data from SFOE [36], June 2023. Fig. 6. DHC system power distribution by main energy source. Air +HP has a 0.002 % value and solar thermal 0 % (not shown). Data from SFOE [36], June 2023. L. Brauchli et al. Energy 333 (2025) 137159 4 methodology of Schluck et al. [28], area specific heat demand on values from Streicher et al. [29] and the Swiss norm SIA 2024 [30]. As the space heating and hot water demands have different temperature requirements, exergy demand is calculated separately using different exergy factors. A range of formulas for exergy factor ε of network-based heat supply have been presented in literature, each with their individual particularities. For a heat flow transferred from/to a fluid that changes the fluid temperature from T1 to T2 with reference temperature Tref , the exergy factor is defined as follows [27]: ε =1−Tref T1−T2 ln(T1 T2)(i) To apply this formula, three key aspects have to be defined first. a) Reference environment Most prominently, there is the question how to define the reference environment, which in this case reduces to the reference temperature. In IEA Annex 49 by Torío & Schmidt [31] four options are examined: he universe (≈0 K), indoor air, undisturbed ground, and ambient air. They recommend using outdoor air as the reference environment for exergy analysis in building energy systems. b) Time-dependency Outdoor temperature fluctuates daily, seasonally, and geographically, making the choice of reference temperature complex. Pons [32] argues that reference temperatures should be chosen constant, while Torío & Schmidt distinguish between steady-state and dynamic approaches [31]. Dynamic approaches consider fluctuating reference temperatures accompanied by fluctuating exergy content, thus capturing storage effects. However, this requires dynamic modelling. Steady-state approaches use average system and outdoor temperatures over a defined period. Torío & Schmidt [33] applied this using local average outdoor temperature during the heating season. Gong & Werner [27] used the yearly average temperature of central northern Europe. This approach is based on both temporal and spatial averaging. In Ref. [31], monthly local averages for the coldest and warmest months were used. The authors concluded that steady-state exergy analysis provides correct insights during exergy performance assessment and is suitable for system comparison. c) Choice of system boundaries As stated by Stephan & Schebek in Ref. [34], “energy and exergy losses are either included or omitted in the results” depending on system boundary selection. Key considerations include. - Demand side: Either the temperature of useful heat is chosen, i.e. room temperature for space heating and tap temperature for domestic hot water. This captures exergy destruction in the distribution system and is suitable for full supply chain analysis. Alternatively, water temperatures on primary or secondary side at the heat transfer station can be used when changes in the building distribution system are not considered. - Supply side: The boundary can either include the energy conversion unit or focus solely on the water temperatures within the flow and Fig. 7. Part of the city of Basel with polygons drawn in QGIS. Red dots indicate buildings connected to a network and coloured areas mark the supply areas of different networks [42]. Dots within polygons form the building cluster supplied by the corresponding DHC network. L. Brauchli et al. Energy 333 (2025) 137159 5 return pipes, depending on whether conversion efficiency is of interest. The approach of Torío & Schmidt [33] was chosen, using local average outdoor temperature during the heating season from October to April for each system. This is consistent with their recommendation in Ref. [31]. The system boundary to quantify exergy demand is drawn to include the water temperatures on the secondary side (Fig. 3). It is assumed that existing building heat distribution systems remain unchanged, keeping temperature demand constant. Exergy loss at the heat transfer station is still considered and decreases with lower primary supply temperatures. The analysis is restricted to the residential and service sector and therefore consists of space heating and domestic hot water demands, excluding process heat. The exergy factor for space heat supply ε SH is calculated using the flow and return temperatures TF and TR of the secondary side according to Ref. [27]: ε SH =1−Tref TF−TR ⋅ln(TF TR)(ii) For each individual building, the flow and return temperatures are estimated based on age according to Table 1. For domestic hot water production, the exergy factor ε DHW is the same but replaces flow temperature with the target hot water temperature of TDHW =60◦C (333 K), stipulated to prevent Legionella growth, and the return temperature with an approximate freshwater temperature of TFW =10◦C (283 K): ε DHW =1−Tref TDHW −TFW ⋅ln(TDHW TFW )(iii) Fig. 8. Process to identify Swiss District Heating and Cooling (DHC) systems. The Register of Buildings and Dwellings (RBD) database contains information on buildings connected to DHC networks. The Swiss Federal Office of Energy (SFOE) database contains information on thermal networks including their location. Depending on the geospatial relations amongst the DHC systems, buildings and networks were clustered and linked either manually or automatically. L. Brauchli et al. Energy 333 (2025) 137159 6 2.3. Grouping of DHC systems The previous sections describe the methodology of characterising districts using the following four KPIs as introduced in section 2.1. (1) Degree of connection (DOC) weighted by exergy demand (2) Exergy demand density (3) Share of residential energy reference area (ERA) (4) Share of single-family houses amongst residential buildings The final step of the methodology consists of grouping similar DHC systems with reference to these KPIs. Inspired by Murray et al. [35] a K-medoid algorithm is used to form groups and identify representative medoids for a set of DHC systems - be it the entire set of all systems or specific subsets. The terms “groups/grouping” are used here to distinguish it from the geospatial “clusters/clustering” of buildings in the following chapter. Unlike geospatial clustering in real-world 2D coordinates, grouping takes place in a n-dimensional KPI space. The algorithm minimizes dissimilarity within each group relative to its medoid, a centrally located point. Fig. 4 illustrates this with an arbitrary 2D example with three groups. The selection of KPIs and number of groups is user-defined and tailored to the dataset. This study uses a 4-dimensional, with each KPI as a dimension and each point representing a DHC system. Groups contain systems with similar KPI profiles, and the medoid represents the most typical system used as a proxy archetype. To ensure uniform weighting, all metrics are equally normalised. Since KPIs (1), (3) and (4) are percentages, it exergy demand density (2) is scaled similarly by normalising it to its maximum variation in the dataset. This methodology is adaptable to any set of DHC systems, and the set can be curated based on the selected boundary conditions such as region or system size. 3. Case study Switzerland 3.1. Energy supply by district heating and cooling networks in Switzerland Switzerland has over 1100 registered DHC systems published in a continuously-updated database of the Swiss Federal Office of Energy (SFOE) [36]. The figures presented from this database are for June 2023. Fig. 5 illustrates network distribution by energy source, revealing that approximately 70 % are biomass-fired. In contrast, Fig. 6 shows distribution by heat supply, indicating that more than 40 % of total heat is provided by Waste-to-Energy (WtE) plants. This disparity highlights the dominance of biomass-fired networks in terms of numbers, and the significant contribution of WtE plants in terms of supply. Fig. 9. Several automatically identified DHC systems. The markers represent network coordinates, which usually correspond to the position of the energy conversion units (taken from the SFOE list); points represent the location of buildings connected to networks (taken from the RBD). Black points represent outliers. Markers and points with the same colour represent a DHC system. L. Brauchli et al. Energy 333 (2025) 137159 7 Table 2 Nine types of spatial relationships amongst clusters, networks, outliers. According to type, different linking strategies are employed to correctly identify DHC systems. Markers represent network location (from SFOE), and points indicate the location of network-connected buildings (from RBD). All markers are blue before linking, and match building colour afterwards, whereby shared colours represent identified DHC systems. The number of identified systems of each type is given, alongside the percentage of all network-connected buildings these systems represent. L. Brauchli et al. Energy 333 (2025) 137159 8 Conventionally, networks utilise fossil fuels for peak load coverage. Biomass-fired networks are of particular interest due to their reliance on a flexible, exergetically valuable, and limited renewable resource. However, Switzerland’s biomass potential is limited and already largely utilised [37,38] and should be prioritised for replacing fossil fuels in high-temperature industrial processes. Waste incinerators supplying buildings suffer from low annual and exergetic efficiency, due to unused summer heat and low temperature demands. 3.2. Data availability in Switzerland The Federal Register of Buildings and Dwellings (RBD) from the Federal Statistical Office provides comprehensive data on Swiss building stock [39] including heating system details to identify DHC-connected buildings. It also includes building type, footprint, and age, useful for estimating usage and energy demand. While cantons manage the database, local communities collect the data, leading to variations in quality and completeness [40]. The aforementioned published list of DHC systems [36] includes network location, energy sources, power, and energy supply. While Table 3 Resulting Medoid DHC systems for the four subgroups distinguished by power range and the energy sources biomass and WtE (Waste-to-Energy). Shaded columns give the KPI values for each Medoid. The other columns provide additional information on the five medoids and the group metrics, including number of systems within the groups, connected buildings, as well as total energy and exergy demand. Medoid Information Group Information Nr. of Buildings Power Exergy Demand Density Degree of Connection Share of residential ERA Share of SFH Nr. of DHC systems Nr. of Buildings Energy Demand Exergy Demand Source District/ Location Power –[MW] [kWh/m2 a] [%] [%] [%] – – [GWh/a] [GWh/a] Biomass <1 MW Wila ZH 31 0.5 3.4 37 83 79 150 4541 172 23 1–10 MW Losone TI 56 7.6 3.0 34 78 67 121 7536 292 38 Murten FR 113 8.0 5.2 31 66 55 80 3949 292 40 >10 MW Pratteln BL 346 15 5.6 54 70 55 19 5170 358 48 WtE Renergia LU 628 46 5.5 52 66 50 28 26 ′ 285 2340 323 Fig. 10. Topology of the first DHC system archetype, automatically identified via algorithm, located in Wila ZH. L. Brauchli et al. Energy 333 (2025) 137159 9 with high-temperature sources. However, this is concomitant with the fact that these systems with their high-exergy sources have the greater potential in terms of increasing exergetic efficiency in the heat supply. The introduction of these archetypes enables more systematic analyses, analogous to the established building archetypes, by using standardised, representative boundary conditions. This approach avoids the limitations of analysing individual DHC systems in isolation, allowing for broader applicability and comparability of results across different studies. Additionally, the use of archetypes facilitates the identification of suitable decarbonisation pathways for specific types of networks that share similar archetype characteristics, thereby enabling the development of generic guidelines that outline recommended measures and pathways for achieving full decarbonisation of existing DHC systems. Despite the strengths of the approach, the qualitative tuning of clustering parameters introduces a level of subjectivity, and the sensitivity of results to input data quality suggests a need for further standardisation and validation, particularly across different national contexts. In future work, the method could be applied to updated datasets that better represent emerging low-temperature systems. CRediT authorship contribution statement Luca Brauchli: Writing – original draft, Visualization, Methodology, Investigation, Data curation, Conceptualization. Núria Duran Adroher: Writing – original draft, Software, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Willy Villasmil: Writing – review & editing, Supervision, Methodology, Conceptualization. Markus Auer: Methodology. Edward Lucas: Writing – review & editing. Philipp Schuetz: Software, Data curation. J¨ org Worlitschek: Supervision, Project administration, Methodology, Conceptualization. Declaration of generative AI and AI-assisted technologies in the writing process During the preparation of this work the authors used ChatGPT and DeepL to generate, translate and linguistically revise text. After using these tools, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication. Funding sources This research was funded by the project “Decarbonisation of Cooling and Heating in Switzerland“ (DeCarbCH) within the research program "Swiss Energy Research for the Energy Transition" (SWEET) of Swiss Federal Office of Energy (SFOE). Declaration of competing interest The authors declare that that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgements The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Glossary BEM Building Energy Modeling DBSCAN Density-Based Spatial Clustering of Applications with Noise DHC District Heating and Cooling DHW Domestic Hot Water DOC Degree Of Connection ERA Energy Reference Area MFH Multy Family Home RBD Register of Buildings and Dwellings SFH Single Family Home SFOE Swiss Federal Office of Energy WtE Waste-to-Energy Data availability Data will be made available on request. References [1] Nussbaumer T, Thalmann S, Jenni A, K¨ odel J. Planungshandbuch fernw¨ arme, 1.2; 2018. [2] Rutz D, et al. Upgrading the performance of district heating networks [Online]. Available: www.upgrade-dh.eu; 2019. [3] Oppermann G, Arnold O, K¨ odel J, Büchler M, Jutzeler M. Leitfaden fernw¨ arme/ Fernk¨ alte [Online]. Available: https://www.fernwaerme-schweiz.ch/fernwaerme -deutsch/Dienstleistungen/Leitfaden.php. [Accessed 4 July 2022]. [4] Hangartner D, K¨ odel J, Mennel S, Sulzer M. Grundlagen und Erl¨ auterungenzu Thermischen Netzen. 2018. [5] Sulzer M, Werner S, Mennel S, Wetter M. Vocabulary for the fourth generation of district heating and cooling. Smart Energy Feb. 2021;1:100003. https://doi.org/ 10.1016/J.SEGY.2021.100003. [6] Werner S. Network configurations for implemented low-temperature district heating. Energy Sep. 2022;254:124091. https://doi.org/10.1016/J. ENERGY.2022.124091. [7] Frederiksen S, Werner S. District heating and cooling. Lund: Studentlitteratur AB; 2013. [8] Lund H, et al. 4th generation district heating (4GDH): integrating smart thermal grids into future sustainable energy systems. Energy Apr. 2014;68:1–11. https:// doi.org/10.1016/J.ENERGY.2014.02.089. [9] Persson U, Werner S. Heat distribution and the future competitiveness of district heating. Appl Energy 2011;88(3):568–76. https://doi.org/10.1016/j. apenergy.2010.09.020. [10] Triebs MS, Papadis E, Cramer H, Tsatsaronis G. Landscape of district heating systems in Germany – status quo and categorization. Energy Convers Manag X Mar. 2021;9:100068. https://doi.org/10.1016/J.ECMX.2020.100068. [11] Kotilainen J, Hellstedt J, Tolvanen H. Determining economic feasibility of supply temperature reduction in existing district heating system through thermohydraulic modelling. Energy 2025;329(January):136781. https://doi.org/10.1016/j. energy.2025.136781. [12] Li X, Yilmaz S, Patel MK, Chambers J. Techno-economic analysis of fifth-generation district heating and cooling combined with seasonal borehole thermal energy storage. Energy 2023;285(June):129382. https://doi.org/10.1016/j. energy.2023.129382. [13] Veyron M, Voirand A, Mion N, Maragna C, Mugnier D, Clausse M. Dynamic exergy and economic assessment of the implementation of seasonal underground thermal energy storage in existing solar district heating. Energy 2022;261(PA):124917. https://doi.org/10.1016/j.energy.2022.124917. [14] Gong M, Werner S. Mapping energy and exergy flows of district heating in Sweden. Energy Proc 2017;116:119–27. https://doi.org/10.1016/j.egypro.2017.05.060. [15] Topal H˙ I, Tol H˙ I, Kopaç M, Arabkoohsar A. Energy, exergy and economic investigation of operating temperature impacts on district heating systems: transition from high to low-temperature networks. Energy 2022;251. https://doi. org/10.1016/j.energy.2022.123845. [16] Tzouganakis P, Fotopoulou M, Rakopoulos D, Romanchenko D, Nikolopoulos N. District heating system analysis and design optimization. Energy 2025;326(April): 136349. https://doi.org/10.1016/j.energy.2025.136349. [17] Chicherin S. Top-down GIS-driven method for configuring the network layout of a 5th generation district heating and cooling (5GDHC) system. Energy 2025;328 (April):136639. https://doi.org/10.1016/j.energy.2025.136639. [18] Vrain M, Dussartre V, Lhuillier N, Girard R, Cover CL. A spatially-explicit method for generating prospective district heating scenarios. Energy 2024;313(October): 133902. https://doi.org/10.1016/j.energy.2024.133902. [19] Pelda J, Holler S, Persson U. District heating atlas - analysis of the German district heating sector. Energy 2021;233:121018. https://doi.org/10.1016/j. energy.2021.121018. [20] Mun´ can V, Mujan I, Macura D, Anđelkovi´ c AS. The state of district heating and cooling in Europe - a literature-based assessment. Energy 2024;304(May). https:// doi.org/10.1016/j.energy.2024.132191. [21] Shen P, Wang H. Archetype building energy modeling approaches and applications: a review. Renew Sustain Energy Rev 2024;199(April):114478. https://doi.org/10.1016/j.rser.2024.114478. [22] Mata ´ E, Sasic Kalagasidis A, Johnsson F. Building-stock aggregation through archetype buildings: france, Germany, Spain and the UK. Build Environ 2014;81: 270–82. https://doi.org/10.1016/j.buildenv.2014.06.013. [23] Alrasheed M, Mourshed M. Building stock modelling using k-prototype: a framework for representative archetype development. Energy Build 2024;311 (October 2023):114111. https://doi.org/10.1016/j.enbuild.2024.114111. L. Brauchli et al. Energy 333 (2025) 137159 16 [24] Dahlstr¨ om L, Johari F, Brostr¨ om T, Wid´ en J. Identification of representative building archetypes: a novel approach using multi-parameter cluster analysis applied to the Swedish residential building stock. Energy Build 2024;303 (November 2023):113823. https://doi.org/10.1016/j.enbuild.2023.113823. [25] Chambers J, Zuberi S, Jibran M, Narula K, Patel MK. Spatiotemporal analysis of industrial excess heat supply for district heat networks in Switzerland. Energy 2020;192:116705. https://doi.org/10.1016/j.energy.2019.116705. [26] Schneider S, Khoury J, Lachal B, Hollmuller P. Geo-dependent heat demand model of the Swiss building stock expanding boundaries : systems thinking for the built environment GEO-DEPENDENT HEAT DEMAND MODEL OF THE SWISS BUILDING STOCK. In: World fustainable built environment conference 2017, conference proceedings (WSBE17); 2017. p. 116–1172 [Online]. Available: https://archive-ouverte.unige.ch/unige:84240/ATTACHMENT01. [27] Gong M, Werner S. Exergy analysis of network temperature levels in Swedish and Danish district heating systems. Renew Energy 2015;84:106–13. https://doi.org/ 10.1016/j.renene.2015.06.001. [28] Schluck T, Streicher KN, Mennel S. Statistical modelling of the energy reference area based on the Swiss building stock. J. Phys. Conf. Ser. 2019;1343:12031. https://doi.org/10.1088/1742-6596/1343/1/012031. [29] Streicher KN, Padey P, Parra D, Bürer MC, Schneider S, Patel MK. Analysis of space heating demand in the Swiss residential building stock: element-based bottom-up model of archetype buildings. Energy Build Feb. 2019;184:300–22. https://doi. org/10.1016/J.ENBUILD.2018.12.011. [30] SIA. SIA 2024 - raumnutzungsdaten für die Energieund Geb¨ audetechnik. Schweizerischer Ingenieurund Architektenverein; 2021. [31] Torio H, Schmidt D. Low exergy systems for high-performance buildings and communities - ECBS Annex 49 Final Report. 2011. [32] Pons M. On the reference state for exergy when ambient temperature fluctuates. Int J Therm 2009;12(3):113–21. https://doi.org/10.5541/ijot.246. [33] Torío H, Schmidt D. Development of system concepts for improving the performance of a waste heat district heating network with exergy analysis. Energy Build 2010;42(10):1601–9. https://doi.org/10.1016/j.enbuild.2010.04.002. [34] Stephan P, Schebek L. Evaluation of district heating systems based on exergy analysis. Darmstadt: University of Technology; 2018. [35] Murray P, Marquant J, Niffeler M, Mavromatidis G, Orehounig K. Optimal transformation strategies for buildings, neighbourhoods and districts to reach CO2 emission reduction targets. Energy Build 2020;207:109569. https://doi.org/ 10.1016/j.enbuild.2019.109569. [36] Bundesamt für Energie BFE. Thermische netze. https://opendata.swiss/de/datase t/thermische-netze-nahwarme-fernwarme-fernkalte. [37] Keel A, Kühne A, Chrenko R. Grundlagenarbeit zu einem Monitoring Holzenergie in der Schweiz 2023:1–64. no. November, https://www.bafu.admin.ch/dam/ba fu/de/dokumente/wald-holz/fachinfo-daten/Schlussbericht_Monitoring_Holzen ergie_23112023.pdf.download.pdf/Schlussbericht_Monitoring_Holzenergie_23112 023.pdf. [38] Spillmann T, Kemmler A, Kreidelmeyer S, Strassburg S, Althaus H-J, Wunderlich A. Energieperspektiven 2050+exkurs biomasse [Online]. Available: www.bfe.admin. ch; 2021. [39] Federal Statistical Office, “Register of buildings and dwellings.” https://public. madd.bfs.admin.ch/. [40] Federal Statistical Office, “Monitoring register of buildings and dwellings”. [41] QGIS Development Team. QGIS geographic information System. Open Source Geospatial Foundation Project; 2023 [Online]. Available: https://www.qgis.org/. [42] Regierungsrat des Kantons Basel-Stadt. Medienmitteilung Regierungsrat: ausbau der Fernw¨ armeversorgung in Basel - 460 Millionen Franken Investitionen für den Klimaschutz [Online]. Available: https://www.bs.ch/nm/2020-ausbau-derfernwaermeversorgung-in-basel—460-millionenfranken-investitionen-fuer-den-klimaschutz-rr.html#:~:text=Schonh eutebetreibtdieIWB,rund75%25CO2-neutral. [Accessed 19 May 2021]. [43] Ester M, Kriegel H-P, Sander J, Xu X. A density-based Algorithm for discovering clusters in large spatial databases with noise. 1996. L. Brauchli et al. Energy 333 (2025) 137159 17