=Paper=
{{Paper
|id=Vol-2541/paper7
|storemode=property
|title=A Vision Towards a Method for Identifying and Mitigating Unsustainable Practices in Organisations
|pdfUrl=https://ceur-ws.org/Vol-2541/paper7.pdf
|volume=Vol-2541
|authors=Letícia Duboc,Raquel Mendonça,Jessica Salvador,Aline Paes
|dblpUrl=https://dblp.org/rec/conf/re/DubocMSP19
}}
==A Vision Towards a Method for Identifying and Mitigating Unsustainable Practices in Organisations==
https://ceur-ws.org/Vol-2541/paper7.pdf
A Vision Towards a Method for Identifying and Mitigating
Unsustainable Practices in Organisations
Leticia Duboc Jessica Salvador
Departament d’Enginyeria Departamento de Informática e Ciência da Computação
La Salle - Universitat Ramon Llull Universidade do Estado do Rio de Janeiro
Barcelona, Spain Rio de Janeiro, Brazil
l.duboc@salle.url.edu jr.salvador.br@gmail.com
Raquel Mendonça Aline Paes
Departamento de Informática e Ciência da Computação Institute of Computing
Universidade do Estado do Rio de Janeiro Universidade Federal Fluminense
Rio de Janeiro, Brazil Niterói, Rio de Janeiro, Brazil
raquelmendonca123@gmail.com alinepaes@ic.uff.br
Abstract—Sustainability is a major concern of our time. Com- financial return is no longer a reason not to embrace sustain-
panies can have a considerable negative impact on the en- ability. Consistent with this view, 85% of them claim to have
vironment, as their productive processes often contribute to incorporated sustainability into their business, even when
the emission of greenhouse gases, generate toxic waste and they cannot quantify the benefits. While the survey observes
consume natural resources. As such, they also have a great that the CEOs are showing a growing understanding of sus-
share of the responsibility towards our sustainable develop- tainable development and are making deeper commitments
ment. This paper presents a vision for a method to identify to solve global challenges, it concludes that there is still
and mitigate unsustainable practices in business organisations. room for improvement.
The method is inspired on the KAOS Framework, in the sense We argue that in order to contribute to our sustainable
that it offers catalogues for the identification and resolution of development, companies need to understand which of their
obstacles to sustainability. The method will be complemented practices negatively affect sustainability and to seek alter-
by a metamodel, a semi-structured language and a knowledge natives or compensations for these practices. We will start
base, to eventually allow automation. from the environmental dimension of sustainability [2] [13]
and, as therefore, we define the following research question:
“How to identify practices in business operations that
1. Introduction negatively affect the environmental sustainability and seek
ways to mitigate these practices?”
The negative impact of human activities are more notice- This paper describes a vision for a method with this very
able than ever. Companies productive processes can often purpose. The method is inspired by the well-known KAOS
contribute to the emission of greenhouse gases, the excessive Framework, a goal-oriented modelling technique in Require-
consumption of natural resources and the generation of ments Engineering [15]. Adapting the goal-obstacle analysis
waste. Sustainability is arguably in the strategic path of in the KAOS Framework, we propose the concepts of “sus-
many companies, but for decades their environmental, so- tainability goal” and “sustainability obstacle” and envision a
cial and governance activities have been disconnected from method that will offer catalogues for identifying and mitigat-
this vision. Many companies still launch ad-hoc initiatives ing sustainability obstacles (i.e. unsustainable practices) in
simply to enhance their “green credentials”, to comply with business organisations. The method will be complemented
regulations or to deal with emergencies; rather than viewing by a sustainability metamodel, a semi-structured language
sustainability as something with a direct impact on their and a knowledge base to eventually support the automatic
business results [4]. identification and resolution of sustainability obstacles. Once
Gradually, many companies are recognising this impact realised, this vision can advance the field of Requirements
and the need to incorporate sustainability practices into Engineering for Sustainability as it will help companies to
their business. A 2016 survey [2] of 1,000 CEOS from 27 identify and mitigate unsustainable goals in their systems
industries across 103 countries found that 89% of CEOs and operations.
believe their commitment to sustainability translates into a The paper is organised as follows: Section 1 motivates
real impact on their industries and the lack of short-term our work; Section 2 summarises the most important concepts
Copyright c 2019 for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
that inspired our vision; Section 3 describes our envisioned A goal is a prescriptive statement of intent that a system
solution; and finally, Section 4 concludes the paper. must meet through the collaboration of its agents. Goals
range from high-level business goals, whose satisfaction
2. Background requires the cooperation of multiple agents (e.g., "Maximize
[Profit]", to low-level technical goals, whose satisfaction
This section briefly explains fundamental concepts re- depends on a single agent (e.g. “Achieve[Calculated Product
lated to our vision: sustainability, goal-oriented requirements Tax]”).
engineering and sustainability models/metamodels. In a goal model, goals are organised into AND / OR
refinement structures. An AND refinement connects a goal
to a subset of goals, i.e. the goal will only be satisfied if
2.1. Sustainability all its children are satisfied. An OR refinement connects
a goal with an alternative set of refinements, i.e. the goal
Sustainability is a complex and interdisciplinary concept will be met if at least one of the refinements is met. The
that has been defined in many ways. Yet, the term can be KAOS Framework draws attention to the fact that often the
essentially understood as the “ability to endure” [1] [3]. goal model is initially developed in an “idealized” way.
Possibly, the most widespread related concept is the one of That is, it fails to consider exceptional conditions in the
sustainable development, defined as the “development that application domain that may violate these goals. To build a
meets the needs of the present without compromising the more realistic model, the KAOS Framework defines goal-
ability of future generations to meet their own needs” [12]. obstacle analysis. This analysis takes a pessimistic view of
These definitions imply that a society that is not sustainable the elaborated model seeking to identify exceptional condi-
cannot be maintained in the long run and will cease to tions (obstacles) and ways to mitigate them (resolutions).
function at some point in time [7]. Therefore, an obstacle is a situation that, if occur, can
Sustainability is composed of several dimensions. The prevent a goal from being satisfied. Thus, every obstacle
three best known are environmental, social and economic, generated needs to be solved in one way or another [10].
often referred as the three pillars of sustainability. However, Critical obstacles are normally resolved through counter-
other dimensions start to be considered; such as the human measure goals, while non-critical ones can be monitored
and the technical [6] [13]. The paper focuses on the environ- or resolved when they occur. In any case, one needs to
mental dimension, which addresses the effects of long-term determine the appropriate resolutions for each obstacle.
human activities on natural systems, including issues related KAOS has an obstacle resolution catalogue that can be used
to ecosystems, climate change, food production and waste, to reduce or mitigate obstacles.
among others.
However, in order for sustainability to be effectively 2.3. Sustainability Models and Metamodels
achieved, other dimensions must also be considered. We
believe that analogous solutions to the one described here There are currently sustainability models that help to
can be developed for the remaining dimensions. incorporate sustainability into organisations and/or systems.
Some of them are summarised below:
2.2. Goal-Oriented Modelling and KAOS Cabot et al. [5] advocate the use of a sustainability
taxonomy combined with goal-oriented techniques. They
Goal-oriented Requirements Engineering (GORE) has explore a preliminary method for modelling and integrating
emerged as an attempt to solve many of the problems sustainability issues in business (in general) and software
of traditional requirements engineering, including incorrect projects (in particular) through the use of the i * Framework.
assumptions about the environment, the little attention given To do so, they explicitly represent the sustainability effect of
to understanding the need for a particular system and to each business or project alternative, thus enabling stakehold-
whether its specifications really captured the needs of stake- ers to understand the commitments between sustainability
holders [9]. Therefore, GORE encompasses the elicitation, and other business goals and to make the best decisions.
assessment, design, structuring, documentation, and analysis Stefan et al. [14] use goal-oriented requirement en-
of software requirements [9]. There are several approaches gineering techniques to help organisations to make more
and methodologies that enable its application. effective decisions for achieving their sustainability goals.
The KAOS is a well-known GORE framework, whose Their method provides systematic techniques for refining
acronym means "Keep All Objectives Satisfied". KAOS is goals into sub-goals, managing goal conflicts, identifying
composed of a modeling language and a method for de- and resolving obstacles to goal achievement, and exploring
veloping software requirements. The most important KAOS and evaluating alternatives to goal achievement.
concepts for our vision are: agent, goal, obstacle, resolution Mahaux, Heymens and Saval [11] use requirements
and goal-obstacle analysis. These concepts are explained engineering techniques to describe requirements that seek
below [15]: minimal environmental impact. Their work aims to provide
An agent is an active component of the system that plays insight into how sustainability requirements can be discov-
a role in meeting a goal. Agents can be humans, devices, ered, what existing tools or techniques can facilitate this
software, etc; they perform operations assigned to them. task, and what their limitations are in this regard.
Penzenstadler and Femmer [13] present a reference to the negated goal. So, an obstacle could be “Delivery
model for software development projects that breaks down car consumes diesel” and a possible resolution could be:
sustainability into its environmental, human, social, eco- “Achieve[Product delivery by bike]”, as shown in Figure 1.
nomic and technical dimensions. The model provides a
series of activities and relates them to the values they support
and the indicators against which they can be evaluated.
Models, such as Cabot et al.’s , Mahaux et al.’s and
Stefan and Letier’s , are specific to the problem they seek to
solve and cannot be used as a reference for business organ-
isations in general. Penzenstadler and Femmer’s model, on
the other hand, seeks to be more generic and, although de-
signed for software development companies, can be adapted
to business processes in general.
3. Envisioned Solution
In order to position themselves as sustainable businesses,
companies need to routinely analyse their own practices to
identify opportunities to become more sustainable. We envi-
Figure 1: Sustainability goal-obstacle analysis.
sion a method to allow companies to recognise unsustainable
practices in business operations and to modify, replace or
compensate them with alternative practices.
3.2. The Method
3.1. Underlying concepts Similar to KAOS, we envision a method that identi-
fies unsustainable practices from a sustainability obstacle
The method is inspired by the goal-obstacle analysis of catalogue and potential mitigation strategies from a sus-
the KAOS Framework. Therefore, we refer to the afore- tainability resolution catalogue. We also envision the semi-
mentioned harmful practices as sustainability obstacles and automation of this process through the definition of a meta-
to the alternative practices as sustainability resolutions. We model, a semi-structured language, rules and a knowledge
extended the KAOS Framework to define these concepts as: base.
A sustainability goal is a prescriptive statement of Therefore, the method will have the following elements,
intent that contributes to a long-term positive impact on whose relationship is represented in Figure 2. These are
one or more sustainability dimensions. That is, just like in described below:
KAOS, a goal is something that needs to be achieved, but in
this case, it also needs to have a long-term positive impact
on one or more sustainability dimensions.
• A sustainability goal is a prescriptive statement of
intent that contributes to a long-term positive impact
on one or more sustainability dimensions. That is,
just like in KAOS, a goal is something that needs
to be achieved, but in this case, it also needs to
have a long-term positive impact on one or more
sustainability dimensions.
• A sustainability obstacle is defined as a precondi-
tion for the non-satisfaction of a sustainability goal.
That is, an obstacle is a situation that occurs that
prevents a goal from being reached.
• Sustainability resolutions are alternatives that can
alleviate or avoid a particular sustainability obsta-
cle.
In a simplified example, a company may have high-level
goals such as "Maximize[Environmental sustainability]" and Figure 2: Elements of the envisioned solution.
lower-level goals like "Reduce[Carbon emissions from prod-
uct delivery]". In KAOS, a goal-obstacle analysis would
first negate the goal (e.g. "NOT Reduce[Carbon emissions • A sustainability metamodel will describe the basic
from product delivery])" to then find obstacles that lead sustainability concepts that underlie the solution.
Therefore, it will be the basis of the sustainability tool; and, finally, (5) the creation of a knowledge base for
obstacle and resolution catalogues, of the language the recommendation tool. We plan to develop this solution
and rules that describes them, and of the knowl- initially for a single domain, starting with a real case study
edge base. We envision that the metamodel will on retail, to then extend to other domains.
have elements such as "goal", "obstacle", "resolu-
tion", "agent", "resource", "environmental impact". 5. Acknowledgement
We intend to extend the work of Penzenstadler and
Femmer [13] to create the metamodel. The research leading to these results has received fund-
• A catalogue of generic sustainability obstacles, ing from the European Union’s Horizon 2020 research
listing possible environmental problems resulting and innovation programme under the Marie Skłodowska-
from common company practices. The catalogue Curie grant agreement No 712949 (TECNIOspring PLUS)
will help to identify the obstacles to sustainable and from the Agency for Business Competitiveness of the
business operations. At least two types of obstacles Government of Catalonia.
will be covered: "pollution" and "resource scarcity".
The former will be concerned with obstacles that References
contribute to different types of pollution (including
water, air and soil), while the latter refers to the [1] The Oxford Dictionary of English. Oxford University Press, 2010.
scarcity of natural resources. Sustainability.
• A catalogue of sustainability resolutions, also [2] Accenture. The un global compact-accenture strategy ceo study 2016
generic, will detail different alternatives to solve or - agenda 2030: A window of opportunity. https://www.accenture.com/
us-en/insight-un-global-compact-ceo-study, 2016.
mitigate sustainability obstacles. The catalogue will
contain, among other things, alternative resolutions [3] C. Becker, R. Chitchyan, L. Duboc, S. Easterbrook, B. Penzenstadler,
N. Seyff, and C. Venters. Sustainability design and software: The
for generic obstacles and concrete examples of such karlskrona manifesto. In Proceedings of the 37th Int’l Conference on
resolutions. Software Engineering-Volume 2, pages 467–476. IEEE Press, 2015.
• A semi-structured language and rules will de- [4] S. Bonini and S Görner. The business of sustainability: Putting
scribe sustainability obstacles and resolutions, as it into pratice. http://www.mckinsey.com/business-functions/
well as rules to automate the resolution search. We sustainability-and-resource-productivity/our-insights/
envision that the language will not only formalise the-business-of-sustainability-mckinsey-global-survey-results",
2011.
the description of sustainability obstacles and res-
[5] J. Cabot, S. Easterbrook, J. Horkoff, L. Lessard, S. Liaskos, and
olutions, but will also make it easier to identify J. Mazon. Integrating sustainability in decision-making processes:
them through the use of a knowledge base. The A modelling strategy. In 2009 31st International Conference on
language will also facilitate the understanding and Software Engineering - Companion Volume, pages 207–210, May
the communication of sustainability obstacle and 2009.
resolutions, as everyone involved understands the [6] Robert Goodland and W. Bank. Sustainability: Human, social, eco-
meaning of the concepts. nomic and environmental. Social Science, 6:220–225, 01 2002.
• A knowledge base will be created for different [7] R. Heinberg and D. Lerch. What is sustainability?, 2010.
industries to allow the automatic identification and [8] Greenpeace International. How clean is your cloud?.
resolution of sustainability obstacles, possibly using http://www.greenpeace.org/international/en/publications/
Campaign-reports/Climate-Reports/How-Clean-is-Your-Cloud/,
predicate logic and Prolog. 2012.
[9] Alexei Lapouchnian. Goal-oriented requirements engineering: An
4. Conclusion overview of the current research. 01 2005.
[10] Emmanuel Letier. Reasoning about agents in goal-oriented require-
Sustainability has become a major concern for the so- ments engineering. 01 2001.
ciety. Large companies like Google, Microsoft, and Apple [11] Martin Mahaux, Patrick Heymans, and Germain Saval. Discovering
have been criticised for not prioritising sustainable practices, sustainability requirements: An experience report. pages 19–33, 03
such as the use of energy efficient data centres [8]. While, 2011.
companies have grown their understanding of and commit- [12] World Commission on Environment and Development. Our common
ment towards sustainability, they can still do more [2]. This future: Report of the world commission on environment and devel-
paper presents a vision for a method inspired in the goal- opment. Technical report, 1987.
obstacle analysis of the KAOS Framework for identifying [13] Birgit Penzenstadler and Henning Femmer. A generic model for
and mitigating unsustainable practices in business organ- sustainability with process- and product-specific instances. In Pro-
ceedings of the 2013 Workshop on Green in/by Software Engineering,
isations. In order to realise this vision, we have planned GIBSE ’13, pages 3–8, New York, NY, USA, 2013. ACM.
the following steps: (1) a more in-depth bibliographic study [14] David Stefan, Emmanuel Letier, Mark Barrett, and Mark Stella-
of the KAOS framework; sustainability models/metamodels; Sawicki. Goal-oriented system modelling for managing environmen-
and recommendation systems. (2) The creation of a data- tal sustainability. 07 2011.
model for the envisioned method; (3) The development of [15] Axel van Lamsweerde. Requirements Engineering: From System
the obstacles and resolution catalogues; (4) the design of Goals to UML Models to Software Specifications. Wiley Publishing,
a semi-structured language, rules and a recommendation 1st edition, 2009.