=Paper=
{{Paper
|id=Vol-3812/paper5
|storemode=property
|title=Requirements and Architectures for Green Configuration
|pdfUrl=https://ceur-ws.org/Vol-3812/paper5.pdf
|volume=Vol-3812
|authors=Andreas Falkner,Richard Comploi-Taupe,Katrin Müller,Sophie Rogenhofer
|dblpUrl=https://dblp.org/rec/conf/confws/FalknerCMR24
}}
==Requirements and Architectures for Green Configuration==
https://ceur-ws.org/Vol-3812/paper5.pdf
Requirements and architectures for green configuration
Richard Comploi-Taupe1,† , Andreas Falkner1,*,† , Katrin Müller2,† and Sophie Rogenhofer1,†
1
Siemens, Vienna, Austria
2
Siemens, Berlin, Germany
Abstract
Green Configuration combines product configuration technologies with environmental impact calculations and enables customers
to balance cost drivers and environmental impact drivers (such as CO2 footprint) for their preferred product variants. We analyse
requirements for configurable products that go beyond the state of the art of classical Life Cycle Assessment (LCA), and we list
corresponding challenges for configurators, such as missing environmental impact data, total costs over the product life cycle, confidence
in data accuracy, performance of the calculation, multi-objective optimisation, comparability of the results, and efficient explanations. To
address those challenges, we discuss three architecture variants which go beyond sequentially calling separate tools for configuration and
LCA: loosely coupled (where the configurator communicates via parameters with the LCA tool), tightly coupled (where the configurator
also manages the basic environmental data and lets the LCA tool calculate the impact values for assemblies), and integrated (where the
LCA calculation is implemented as part of the configurator). We find that all architectures rely on complete and reliable input data
(which might be synthesised offline by data-driven AI methods) and have different advantages and disadvantages concerning efforts for
tool vendors, product modellers, and customers.
Keywords
product configuration, sustainability, green configuration
1. Introduction with data from usage and end-of-life phases can genuine
circularity and optimised sustainability (e.g., maximising
With the European Green Deal [1, 2], the European Union the product’s utility while minimising waste) be reached for
drives the EU society to a more sustainable future. The EU configurable products.
Agenda 2050 defines environmental, economic, and social The term “Green Configuration” was established a few
goals to be achieved by production systems [3]. Requests years ago1 for this enhancement of configuration tools with
for Proposal (RFPs) and other B2B offers of all manufactur- environmental impact calculations. This gives the user com-
ing companies will soon require proof of highly sustainable prehensive information about the specific effects of their
production and operations – due to higher awareness of decisions. Small changes in configuration can have a sig-
customers and national authorities, and stricter laws such nificant impact on the ecological footprint. Multi-objective
as the forthcoming Ecodesign for Sustainable Products Reg- optimisation strategies make it possible to optimise the prod-
ulation (ESPR) [4] or Sustainable Products Initiative (SPI) of uct configuration according to desired dimensions (financial
the EU. and sustainable) depending on specific requirements. Fur-
To persist, companies need to document the Product Car- thermore, provisions must be made so that the final product
bon Footprint (PCF) or even Product Environmental Foot- remains in accordance with the increasingly complex legal
print (PEF) of all their products transparently and reliably, framework. This affects not only sales configurators (where
according to valid or forthcoming regulations like the Digital customers shall see the expected environmental impact and
Product Passport (DPP) [5]. For mass production, processes corresponding costs at the point-of-sale, i.e., before they
to assess the environmental impact have already been de- order a product) but is also vital for engineering configura-
fined and standardised, e.g., Life Cycle Assessment (LCA) is tors (which need to prove that the finally manufactured and
standardised by ISO 14040 [6]. deployed product keeps the promises of the sales phase to
Product configuration [7] and Industry 4.0 architectures avoid penalties or non-compliance costs).
[8] go beyond mass production, and mass customisation Wiezorek and Christensen [11] have given a good
allows to manufacture individualised (i.e., lot-size 1) prod- overview of the topic, and we will extend their work based
ucts. The transition towards a circular economy, as required on the current developments, e.g., by considering various
by ESPR, puts challenges to mass customisation and con- types of environmental impact (not only CO2 equivalents)
figuration systems, such as the promotion of circularity- and by integrating the total cost of ownership (TCO) over
based business models, integration of eco-design princi- the complete life cycle (not only the production phase). Our
ples to serve sustainable business demands (i.e., green pro- goal is to find alternative architectures for combining config-
curement), and documentation and understanding of the uration and environmental impact calculation and evaluate
product’s material characteristics, manufacturing processes, them w.r.t. user requirements and challenges of their appli-
energy usage, and environmental impacts over the com- cation in practice.
plete life cycle. Only by integrating pre-manufacturing data In the next section, we will analyse the state of the art of
environmental data and impact calculation. In section 3, we
ConfWS’24: 26th International Workshop on Configuration, Sep 2–3, 2024,
Girona, Spain discuss which challenges arise when this is to be applied
*
Corresponding author. to configurable products. In section 4, we present the main
†
These authors contributed equally. architectures for green configuration and describe how they
$ richard.taupe@siemens.com (R. Comploi-Taupe); deal with those challenges. Finally, we conclude what this
andreas.a.falkner@siemens.com (A. Falkner); can mean for configurator vendors.
katrin.km.mueller@siemens.com (K. Müller);
sophie.rogenhofer@siemens.com (S. Rogenhofer)
0000-0001-7639-1616 (R. Comploi-Taupe); 0000-0002-2894-3284
(A. Falkner) 1
The term “Green Configuration” has been used more by CPQ solution
© 2024 Copyright for this paper by its authors. Use permitted under Creative Commons License
Attribution 4.0 International (CC BY 4.0). providers than in academia, e.g., by encoway [9] and CAS [10].
CEUR
ceur-ws.org
Workshop ISSN 1613-0073
Proceedings
2. Environmental impact assessment: components from thousands of suppliers across multiple
industrial sectors, which poses considerable challenges to
The state of the art data management and performance.
Life Cycle Assessment according to ISO 14040 [6] has been As PCRs and PSRs try to harmonise the environmental
the means of choice for environmental impact assessment impact assessment within one product category, large-scale
for products, processes, and solutions for decades. More systems such as rolling stock, production lines, or process
and more LCA tools and databases are available, and LCA technology are composed of products or assemblies with
results are used for Environmental Product Declarations multiple PCRs and PSRs to be applied, which are not nec-
(EPDs) according to ISO 14025 [12]. Examples for commer- essarily comparable. Inline environmental assessments are
cial providers are SimaPro2 , iPoint3 , and sphera4 . Ecoin- required independently of PCRs and PSRs, especially in
vent5 is an extensive database used by providers such as large-scale system configuration or turnkey projects. Fo-
SimaPro. Some tools and databases target single environ- cusing on customer-specific usage conditions will provide
mental indicators only – e.g., SiGREEN [13]. ESTAINIUM6 tailored results. However, small changes in the conditions
is an open network to exchange PCF-related data in a non- may significantly impact the product’s LCA results.
profit-oriented way. There is little related work concerning combining LCAs
The LCAs are based on the product’s Bill of Materials with a dynamic modelling approach to consider customer-
(BOM) and Bill of Processes (BOP) along its life cycle. Most specific usage or to adapt the background database to future
LCAs are done after the final product design when the ma- scenarios (cf. Udriot et al. [16] for one example). Such a
terials and processes are identified. LCA can also be applied scenario analyser often applies the same product configu-
earlier in the design process of configurators to improve ration to multiple usage scenarios. Changes in the product
design decisions before finalisation. configuration could be made iteratively and sequentially.
In customer communication, EPDs are often used to show Other research reports about work on guidance to inte-
the results of the LCA. However, the EPDs are based on a grate LCAs in general and EPDs in particular into config-
specific, fully specified product or – less individually and less urators and its evaluation in the construction sector [17].
precisely – on a representative product, an average (fictive) Wiezorek and Christensen [11] suggest an architecture for
product, or the worst-case product of a homogenous product integrating LCA into a configurator based on a profound
family. Thus, they cannot help customers decide on product analysis of sustainability assessments according to the Eco-
details or with customer-specific optimisation. In the best logical Scarcity Method (ESM) and data from the ecoinvent
case, they can give a rough orientation based on existing database – focusing on the supply chain and manufacturing
LCAs for product representatives or typicals. phase and mapping all impact to PCF values. A qualita-
As EPDs are used for customer communication, Prod- tive study [18] lists several advantages that sustainability-
uct Category Rules (PCR) and Product Specific Rules (PSR) focused configurators can potentially provide.
are defined to provide comparable results [12]. PCRs and
PSRs harmonise the system boundaries and provide default
parameters for EPDs. However, the usage scenario to be
3. Challenges of impact assessment
applied in EPD refers to a fictive reference service time. for configurable products
This reference service defines the years of service, load, and
operating hours for calculation purposes only. It does not Manufacturing companies need to document not only the
consider the customer-specific usage conditions. Although PCF but also the PEF (i.e. more environmentally critical sub-
PCR and PSR aim to provide comparable EPD results within stances than just CO2) of all their products transparently
a product category, customers still have to make an effort and reliably, according to valid or forthcoming regulations
to relate these results to the individual life cycle conditions like DPP. This must be based on information from suppliers
and make the right purchase decisions. and knowledge about production processes and operations
Besides the insufficient consideration of the customer- (i.e., usage and end-of-life phases) and includes the selection
specific usage scenario, the broad range of existing back- of suppliers and processes which minimise the overall envi-
ground data sets makes it hard to figure out the product- ronmental impact. In addition, economic key performance
specific environmental performance, as this is influenced indicators (KPIs), such as costs for production, transport,
by applied LCA data sets as well. The LCA data sets of- usage, disposal, etc. need to be considered and require multi-
ten provide a market average or a representative example objective optimisation with good user guidance (including
process and do not reflect a specific supplier’s product and understandable explanations).
production-specific environmental impacts. There is still a The configuration of such an environmentally conscious
gap in using primary data along the supplier chain. system is difficult, especially for complex products, because:
Recent initiatives7 target the PCF accounting and man- • Many suppliers are involved, among them many
agement to improve the primary PCF data share in product small and medium-sized enterprises (SMEs), which
accounting and to provide trusted and reliable data along often cannot provide sufficiently good documenta-
the supply chain. However, even for PCF, several standards tion on materials and PEF (e.g., several thousand
and guidelines are in place [14, 15] – and sufficient methods suppliers for parts of metro trains).
are not yet available to make the data comparable. Large-
• Parts have entirely different properties as they come
scale products may require data on millions of materials and
from different industries such as electrical, engineer-
2
https://simapro.com/ ing, or building technology and may interpret envi-
3
https://www.ipoint-systems.com/ ronmental KPIs differently.
4
https://sphera.com/ • Different countries have a wide variety of regula-
5
https://ecoinvent.org/
6
https://www.estainium.eco
tions and certificates (which may even change over
7
Initiatives such as the aforementioned SiGREEN and ESTAINIUM.
time), so different solutions (i.e., combinations of a day) and energy mix (e.g., how much fossil, how
components) are necessary. much wind power or photovoltaic) [19].
• The environmental impact (e.g., concrete PEF values) 3. Complexity of PEF calculation: The calculation of
depends on the production technologies and loca- the complete product’s environmental impact (e.g.,
tions of the suppliers and the location of deployment CO2 emissions) is more complicated than just adding
and conditions at customer sites. the corresponding values of all the parts [20]. LCA
• Sustainability data for many components is missing tools such as Green Digital Twin™ (GDT) [21] or
or questionable, and improvement is difficult as it is SimaPro implement such details and are certified to
out of the control of the system integrator. comply with the standards.
• The system configuration is often not yet defined in 4. Confidence in calculated data: As the input data
sufficient detail at the offering time, and therefore, come with a certain uncertainty, we must hand over
the environmental impacts can only be estimated that uncertainty to the intermediate and total val-
but not precisely calculated. ues (e.g., with a confidence level or a value range).
• Adaptations during contract negotiations or after de- Plausibility checks (e.g., assembly cannot have less
ployment can affect compliance and/or performance impact than the sum of parts) would be helpful.
and require efficient re-calculation and updating of 5. Multi-objective optimisation: For the customer, it is
documentation. helpful to know about the impact distribution over
the phases (supply chain, production, deployment,
To handle those requirements, we need algorithms and usage, end-of-life) and separately for different im-
techniques for: pact types (energy consumption, pollution, etc.). The
• Calculation of all relevant sustainability metrics at corresponding costs (especially TCO) over different
point-of-sale: This is not possible in advance (as cur- expected lifetime periods (e.g., 10 years vs. 20 years)
rently done) because it depends on user decisions, are vital for good decisions. This means the values
which can lead to billions of potential variants. It for all those metrics must be tracked individually.
must be fast enough to ensure a good user experi- 6. Effective explanations and user guidance: It is insuf-
ence and, therefore, requires high performance. ficient to simply show the user the resulting LCA
• Reliable aggregation of the values of all sub-parts: and TCO values. The user must also understand the
This includes highly accurate approximations for causes for those values, i.e., which of their decisions
missing values specific to the current customer se- contributed most. Transparency must be established
lections. For the usage phase, this cannot be based to support users in understanding the impact of a
on sub-parts alone (as is currently done) but on the specific configuration on economic and PEF KPIs.
functionality of the whole product or sub-systems. 7. Comparability of data: Data often depends on as-
• Guided optimisation of several objectives: It is not sumptions (such as those mentioned in challenge 2),
sufficient to calculate only one (combined, weighted) and players may use different assumptions. To make
optimum (as in current tools). The user must be sup- offers from different vendors comparable, those as-
ported in evaluating the Pareto front efficiently and sumptions and the algorithms used must be dis-
finding the best compromise for conflicting goals. closed or harmonised, e.g., according to standards
• Concise visualisation of the results: This helps the such as ISO 14040 [6].
user to easily understand the impacts of their de-
cisions. It shall explain the system’s confidence in 4. Comparison of architectures for
its calculations and where to change a decision to
achieve a better result (which goes beyond the capa- green configuration
bilities of current systems).
This section presents architectures with increasing degrees
In the remainder of the text, we will focus on the following of integration, starting with simply using an existing con-
concrete challenges of Green Configuration: figurator and feeding its results into an existing or newly
customised LCA calculator. In subsections, we will discuss
1. Missing environmental data from suppliers: Many,
how each deals with the challenges from the previous sec-
especially smaller companies, do not yet disclose
tion and summarise the whole section in a table at the end.
environmental data for their products (partly be-
cause they do not know them themselves). This not
only concerns the supply chain, i.e., the impact of 4.1. Status quo: Separate tools
the production of those sub-parts, but also their us-
A naïve approach to Green Configuration is sequential –
age and end-of-life processing. To ensure proper
based upon the availability of two separate tools: config-
LCA calculation, missing data must be synthesised
urator and LCA calculator. For the configurator (i.e., the
as accurately as possible, i.e., by specific approx-
left lane in Figure 1), a modeller defines the product model
imations based on machine learning from similar
(i.e., variety and dependencies) in a knowledge base (KB)
suppliers and/or components, simulation of produc-
by using the integrated development environment (IDE) for
tion and/or operation, using intelligent extrapola-
the configurator. A customer or salesperson uses the con-
tion which takes trends into account (e.g., new ver-
figurator user interface (UI) to set values to configuration
sions of components typically get better).
parameters to fulfil their requirements. Continuously, the
2. Unclear impact data for the usage phase: The envi- solver checks compliance with the KB and sets other param-
ronmental impact is customer- and even application- eters accordingly. Only when the configuration is finished,
specific. It depends on the context, such as operating the solver hands over the resulting BOM to the LCA tool
hours (e.g., whether an engine runs 8 or 24 hours
Customer Customer
Configurator LCA tool Configurator LCA tool
UI UI UI
Configuration LCA values Configuration LCA values
values (separate for values (separate for
(incl. costs) phases) (incl. costs) phases)
Solver / LCA Solver / LCA
Optimiser calculator Optimiser calculator
KB LCA data KB LCA data
(variants, (materials, pro- (variants, (materials, pro-
constraints) duction, usage) constraints) duction, usage)
Model UI (for Model UI (for
editor materials) editor materials)
IDE IDE
Modeller Modeller
Figure 1: Sequential architecture Figure 2: Loosely coupled architecture
(right lane). A (typically other) modeller now collects all LCA concepts and continuously maintain this interface to
necessary LCA data for the materials for all relevant phases comply with the evolving versions of both the configurator
(supply chain, production, usage, end-of-life) and calculates and the LCA tool and API. Modellers need much expertise
the environmental impact values for the product. and additional effort because they must specify the LCA
We will not go into more detail because this approach does model separately from the configurator model and make
not really combine the two tools and is impractical due to sure that both are in sync (i.e., define the core structure and
the typically long duration of the manual LCA assessment. the dynamic parameters, include all relevant materials and
components, map those included components to configu-
4.2. Loosely coupled architecture ration features, i.e., parameters). They may even need to
involve a tool specialist, at least for the first setup of the
To achieve faster results for the user, one can automate the system. The configurator users benefit from the proven
process. Such a loosely coupled approach was taken by, e.g., LCA processes and the typically up-to-date data in the cor-
Tacton [22]. It is based upon modelling the environmental responding databases (e.g., ecoinvent). On the other hand,
impact in an LCA tool (such as SimaPro) and synchronising user experience may still be weak because of possibly long
it with the configurator by mapping configuration features response times in interactive use (due to the overhead of
with parameters for the LCA (as sketched by the dashed calling an external tool and – especially for the first calls –
line between KB and LCA data in Figure 2). After each user the comparably long time to calculate the resulting value).
action in the configurator UI, the LCA calculator is called Optimisation is challenging as the configurator cannot eas-
and returns the adjusted sustainability values to be shown ily access intermediate values for sub-assemblies, thereby
in the configurator UI. The final LCA values may be used steering optimisation in the right direction. This loosely
for optimisation, i.e., minimisation of environmental impact, coupled architecture covers the challenges from section 3
in the configurator (indicated by the dashed arrow from the in the following way:
LCA values to the solver).
The main challenge for the configurator vendor is to de- 1. Missing environmental data from suppliers: Avail-
fine a clean generic mapping between configuration and able LCA data for the sub-parts (from suppliers),
for the manufacturing tasks (in the own production
process), for various time periods in the operations
phase (depending on details of usage and surround-
ings), and for end-of-life (e.g., recycling efforts) can Customer
be reviewed and – if necessary – extended by the
modeller in the LCA tool’s UI before the configu- Configurator LCA tool
ration process starts. Additionally, an external tool
based on machine learning could help to synthesise UI
data offline (this needs to be implemented by other
experts).
2. Unclear impact data for usage phase: Information
Configuration LCA values
about expected usage can be collected as configura-
values (separate for
tion data and handed over as parameters to the LCA
calculator to achieve customer-specific values. (incl. costs) phases)
3. Complexity of PEF calculation: The LCA calculator
can be trusted to comply with the rules for proper
calculation (PCR, PSR).
4. Confidence in calculated data: Current LCA tools do Solver / LCA
not (yet) sufficiently inform about (missing) accuracy Optimiser calculator
of values.
5. Multi-objective optimisation: LCA values of sub-
parts and sub-assemblies are not available to the
optimiser, which can lead to weak (sub-optimal) per- KB LCA data
formance. (variants, (materials, pro-
6. Effective explanations and user guidance: The con- constraints) duction, usage)
figurator UI cannot access the internals of LCA cal-
culation and thus cannot assist the user with expla-
nations and recommendations.
7. Comparability of data: The LCA tool is typically Model /
certified. Therefore, the resulting LCA values are LCA editor
comparable to other calculations based on the same
standards. IDE
4.3. Tightly coupled architecture
Some LCA tools, e.g., Green Digital Twin™ (GDT) from Modeller
Siemens, are generic and expect that the LCA data for the
Figure 3: Tightly coupled architecture
LCA calculation is handed over at the call. This can be
used for a tightly coupled architecture, where the configu-
rator manages the LCA data and just calls the LCA tool (see
Figure 3). of values, but as the solver has access to the LCA
Again, the advantage for the customer is that they are values of sub-assemblies, it can partly validate them.
facing just one UI (for configuration and LCA values). But 5. Multi-objective optimisation: The optimiser can
now, the same is true for the modeller (a single UI for config- use the LCA values of sub-assemblies for informed
uration and LCA models). This means that the configurator heuristics.
vendor must supply such a modelling UI, which allows the 6. Effective explanations and user guidance: The con-
binding of configuration variants to their LCA data (typ- figurator UI cannot access the internals of LCA calcu-
ically extracted from LCA data sets), and a solver which lation but can use the LCA values of sub-assemblies
hands the LCA data for the selected variants over to the for some recommendations.
LCA calculator. The LCA calculator can even be called for 7. Comparability of data: Similar to the loosely coupled
parts of the product (not only for the whole product). The approach, the LCA values are comparable to other
tightly coupled approach covers the challenges from sec- calculations based on the same standards.
tion 3 in the following way:
1. Missing environmental data from suppliers: Simi- 4.4. Integrated architecture
larly to the loosely coupled approach, LCA data for One can go one step further and directly integrate LCA cal-
the relevant sub-parts can be prepared or synthe- culation into the configurator by extending the modelling
sised offline. environment (IDE) with a component for LCA and calcu-
2. Unclear impact data for usage phase: The configura- lating sustainability values directly in the configurator (see
tor hands those LCA data over to the LCA calculator, Figure 4). Such an approach was taken by, e.g., CAS Merlin
corresponding to the customer’s expected usage. [11, 23].
3. Complexity of PEF calculation: The LCA calculator The integrated approach has the advantage that it does
can be trusted to comply with the rules for proper not need an explicit mapping to an LCA tool during mod-
calculation (PCR, PSR). elling and can use environmental data during reasoning
4. Confidence in calculated data: Current LCA tools do and optimisation to come up with a more preferred solu-
not (yet) sufficiently inform about (missing) accuracy tion. On the other hand, it needs considerable effort for the
calculated LCA values for assemblies if the accuracy
of the input data is known or can be estimated.
Customer 5. Multi-objective optimisation: As the optimiser and
LCA calculator are fully integrated, intermediate
Configurator
LCA values can efficiently control optimisation.
UI 6. Effective explanations and user guidance: The com-
plete integration of the solver and LCA calculator
and full access to all their intermediate data allows
for detailed explanations and recommendations.
Configuration LCA values 7. Comparability of data: The extension of the solver
values (separate for with LCA calculation leads to highly individualised
(incl. costs) LCA values. If the configurator vendor does not
phases)
achieve certification (e.g., due to high costs and/or
efforts), the LCA values may not be comparable to
commercial LCA tools.
Solver / Optimiser /
LCA calculator 4.5. Summary
Summing up, all three approaches have strengths and weak-
nesses when dealing with the challenges. Challenge 1 (miss-
ing data) is not discriminating, and the best way to cover
KB LCA data it is by extending and/or improving input data offline, e.g.
(variants, (materials, pro- with the help of data-driven AI. Therefore, we rate only
constraints) duction, usage) challenges 2 to 7 in Table 1 and use a three-valued scale –
the approach has strengths, is neutral, or has weaknesses –
to condense the arguments from the preceding subsections.
Model / Table 1
LCA editor Concerning the challenges, the architectures have strengths (+),
are neutral (o), or have weaknesses (-)
IDE
Loosely Tightly
Challenge Integrated
coupled coupled
2 - usage phase o + +
Modeller 3 - calculation + + o
4 - confidence - o +
Figure 4: Integrated architecture 5 - optimisation - o +
6 - explanations - o +
7 - comparability + + o
configurator vendor to implement the calculation, care for
certification (for LCA calculation according to ISO 14040, The integrated approach offers more value to the cus-
for EPD generation according to ISO 14025), and continu- tomers, e.g. more optimisation possibilities and better ex-
ously maintain it to keep compliance with standards up to planations. On the other hand, this requires more effort for
date. Development efforts can be reduced if certification the configurator developer because they must implement
is unnecessary, e.g., because customers need not compare LCA calculations (not just call existing tools or libraries) and
their products with competitors but only with their internal care for the necessary certification to make the calculations
variants. The integrated approach covers the challenges transparent and comparable.
from section 3 in the following way: The coupled approaches take advantage of re-using off-
the-shelf LCA calculators and can even hand over configura-
1. Missing environmental data from suppliers: Simi- tion information as parameters, but neither (especially the
larly to the coupled approaches, LCA data for the loosely coupled architecture) can easily integrate the calcu-
relevant sub-parts can be prepared or synthesised lation results into their reasoning (e.g. for optimisation and
offline. explanations). The tightly coupled architecture can access
2. Unclear impact data for usage phase: The combined values from sub-assemblies to achieve better usability.
solver and calculator can directly access the expected A product modeller may prefer the tightly coupled ap-
usage information as specified by the customer to proach and especially the integrated approach because data
compute the LCA values. management can be done with only one tool: the configura-
3. Complexity of PEF calculation: Simple impact calcu- tor.
lations (e.g., the addition of upstream) can be easily
integrated into the solver. Covering the same func-
tionality as an LCA tool and achieving certification 5. Conclusions
requires much more effort by the configurator ven-
Green Configuration, the combination of product configura-
dor.
tion technologies with environmental impact calculations,
4. Confidence in calculated data: The combined solver
is a vital approach to address sustainability challenges. We
and calculator can keep track of the accuracy of the
have analysed requirements and challenges and discussed and circular european economy?, 2022. URL: https://
several architectures for configurators implementing a green circulareconomy.europa.eu/platform/en/knowledge/
configuration approach. digital-product-passport-ticket-achieving-climate-
We have seen that the different architectures have dif- neutral-and-circular-european-economy, accessed
ferent strengths and weaknesses, advantages and disadvan- 2024-07-19.
tages. All of them are feasible and require different efforts [6] International Organization for Standardization, ISO
from stakeholders, i.e., tool vendors, product modellers, and 14040:2006 Environmental management — Life cycle
customers. From the viewpoint of a product owner, the se- assessment — Principles and framework, 2006.
lection of their individually preferred architecture depends [7] A. Felfernig, L. Hotz, C. Bagley, J. Tiihonen (Eds.),
on the product‘s complexity, the level of product customisa- Knowledge-based Configuration: From Research to
tion, the number of offers per year, the LCA impact of the Business Cases, 1st ed., Morgan Kaufmann Publishers
usage phase, and the need to enhance customer experience Inc., San Francisco, CA, USA, 2014.
and operational efficiency. [8] H. Lasi, P. Fettke, H.-G. Kemper, T. Feld, M. Hoffmann,
There is much room for future research on efficiently Industry 4.0, Business & Information Systems Engi-
merging sustainability management with configuration life- neering 6 (2014) 239–242. doi:10.1007/s12599-014-
cycle management, e.g., reference architectures, reliable data 0334-4.
exchange, individualised impact calculation, multi-objective [9] S. Keinitz, Corporate sustainability – how
optimisation, elaborate standards, etc. Green Configuration can help!, 2023. URL: https:
As one of the most important, we see the monetary as- //www.encoway.de/en/blog/green-configuration/,
sessment of PEF as a means of providing an estimate of accessed 2024-07-19.
the TCO. Visualising the monetary impact of configuration [10] R. Wiezorek, CPQ-Software: Green Config-
decisions over the whole lifecycle of the product will create uration für mehr Klimaschutz, 2022. URL:
a real incentive for the customer to choose the more sustain- https://www.digital-engineering-magazin.de/
able product configuration (e.g., less energy costs during cpq-software-green-configuration-fuer-mehr-
the operation phase). Green Configuration extended with klimaschutz-a-dc586d2681e3701d3088e0a405ee6185/,
TCO minimisation can lead to a triple-win situation: min- accessed 2024-07-19.
imised total cost of ownership for the customer, increased [11] R. Wiezorek, N. Christensen, Integrating sustainabil-
demand for high-quality products for the industry, and less ity information in configurators, in: M. Aldanondo,
environmental damage. A. A. Falkner, A. Felfernig, M. Stettinger (Eds.), 23rd
Green Configuration enables the creation and scale of International Configuration Workshop, volume 2945
application-specific EPDs and DPPs based on more precise of CEUR Workshop Proceedings, CEUR-WS.org, 2021,
information and assumptions on the concrete product prop- pp. 65–72. URL: https://ceur-ws.org/Vol-2945/52-RW-
erties and usage. To make such specifically customised ConfWS21_paper_16.pdf.
values comparable between tools, existing standards like [12] International Organization for Standardization, ISO
ISO 14040 and the ISO 14020 series [24] need to be adapted 14025:2006 Environmental labels and declarations —
or extended. Transparency of the individual impact values Type III environmental declarations — Principles and
per phase and/or criterion is necessary for well-founded procedures, 2006.
decisions. [13] Siemens, Decarbonization starts with data, 2024.
Disclaimer: Much of the presented work is “thought URL: https://www.siemens.com/global/en/company/
work”. Currently, we are working on prototypes to con- sustainability/product-carbon-footprint.html,
firm the ideas and results in practice. accessed 2024-07-19.
[14] World Business Council for Sustainable Development,
Pathfinder framework – guidance for the accounting
References and exchange of product life cycle emissions, 2021.
URL: https://www.wbcsd.org/resources/guidance-for-
[1] European Commission, The European Green Deal:
the-accounting-and-exchange-of-product-life-cycle-
Striving to be the first climate-neutral continent, 2020.
emissions/, accessed 2024-07-19.
URL: https://commission.europa.eu/strategy-and-
[15] Together for Sustainability, The product carbon foot-
policy/priorities-2019-2024/european-green-deal_en,
print guideline for the chemical industry, 2024. URL:
accessed 2024-07-19.
https://www.tfs-initiative.com/app/uploads/2024/
[2] European Commission, The European Green Deal,
03/TfS_PCF_guidelines_2024_EN_pages-low.pdf,
2019. URL: https://eur-lex.europa.eu/legal-content/
accessed 2024-07-19.
EN/TXT/?uri=COM%3A2019%3A640%3AFIN, ac-
[16] M. Udriot, K. Treyer, O. Buhler, L. Etesi, E. David,
cessed 2024-07-19.
V. Girardin, Rapid life cycle assessment soft-
[3] S. Muench, E. Stoermer, K. Jensen, T. Asikainen,
ware for future space transportation vehicles de-
M. Salvi, F. Scapolo, Towards a green & digital future,
sign, in: Aerospace Europe Conference 2023, 2023.
2022. doi:10.2760/54.
doi:10.13009/EUCASS2023-015.
[4] European Commission, Proposal for ecodesign for
[17] I. Campo Gay, L. Hvam, A. Haug, Automation of life
sustainable products regulation, 2022. URL: https:
cycle assessment through configurators, in: Z. Anišić,
//environment.ec.europa.eu/publications/proposal-
C. Forza (Eds.), 10th International Conference on
ecodesign-sustainable-products-regulation_en,
Mass Customization and Personalization, 2022, pp.
accessed 2024-07-19.
19–25. URL: https://mcp-ce.org/wp-content/uploads/
[5] T. Götz, H. Berg, M. Jansen, T. Adisorn, D. Cembrero,
2022/10/5.pdf.
S. Markkanen, T. Chowdhury, Digital Product
[18] I. Campo Gay, L. Hvam, Sustainability-focused
Passport: The ticket to achieving a climate neutral
product configurators benefits and expectations: A
construction industry case, in: IEEE Interna-
tional Conference on Industrial Engineering and En-
gineering Management, IEEE, 2023. doi:10.1109/
ieem58616.2023.10406559.
[19] H. Ritchie, P. Rosado, Energy mix, Our World in Data
(2020). URL: https://ourworldindata.org/energy-mix.
[20] European Commission, Product Environmental
Footprint Category Rules Guidance, version 6.3,
2018. URL: https://eplca.jrc.ec.europa.eu/permalink/
PEFCR_guidance_v6.3-2.pdf, accessed 2024-07-19.
[21] Siemens, Finding the right balance between
costs and the carbon footprint, 2024. URL:
https://resources.sw.siemens.com/en-US/case-
study-siemensag, accessed 2024-07-26.
[22] Tacton Systems AB, Tacton CPQ environ-
mental footprint configuration, 2024. URL:
https://www.tacton.com/products/tacton-cpq/
environmental-footprint-configuration/, accessed
2024-07-19.
[23] CAS Software AG, GreenConfiguration mit CAS Mer-
lin CPQ, 2024. URL: https://www.cas.de/infothek/
wissenswertes/cpq-wissen/green-configuration/, ac-
cessed 2024-07-19.
[24] International Organization for Standardization,
ISO 14020:2022 Environmental statements and
programmes for products — Principles and general
requirements, 2022.