Ontological Framework for Integrating Environmental Issues within

Sustainable Enterprise

Enhancing Enterprise Decision-making

Edrisi Muñoz

, Elisabet Capón

, José M. Laínez

, Antonio Espuña

and Luis Puigjaner

Centro de Investigación en Matemáticas, Jalisco S/N, 36240, Guanajuato, Mexico

Institute for Chemical and Bioengineering, ETH Zürich, 8093 Zürich, Switzerland

School of Chemical Engineering, Purdue University, West Lafayette, IN, U.S.A.

Department of Chemical Engineering, Universitat Politècnica de Catalunya, Av. Diagonal, 647, E08028 Barcelona, Spain

Keywords:

Environmental modeling, Ontology Reuse, Enterprise Model, Decision Support Systems, Ontology Frame-

work.

Abstract:

Ontologies stand for an excellent choice for building complex models while allowing a high level of ﬂexibil-

ity, re-usability, usability and easiness of maintenance. This work proposes the re-use of an ontological model

for the integrated enterprise in order to include the environmental assessment function. Since enterprises are

complex systems involving different functionalities, decision-making becomes a highly challenging task, and

decision process is usually separated in several levels. Nevertheless, such levels are closely related by the

sharing of data and information. Therefore, effective integration among the different hierarchical levels, by

means of tools which improve information sharing and communication, may play a crucial role for the en-

hanced enterprise operation, and consequently for fulﬁlling the enterprise’s goals. The ontological framework

provides a common modeling framework which facilitates integration among the different decision levels,

and works as the mechanism for supporting information and knowledge sharing among multiple applications.

The general semantic framework developed is applied to a case study comprising an enterprise supply chain

network design-planning problem which considers environmental issues.

1 INTRODUCTION

Currently, enterprises face stricter regulations related

to safety and environmental issues, new processes and

materials, and an increasing degree of uncertainty in

global economics. Even more, companies pursue to

create value in a more sustainable way to fulﬁll their

operations and goals, which implies properly harmo-

nizing proﬁtability and natural resources and energy

consumption (Grossmann, 2004). Therefore, it is nec-

essary to develop and use approaches and tools ca-

pable of analyzing and understanding the processes’

effects on the environment. To succeed in reaching

such a balance, environmental performance must be

integrated into the business strategy and development.

Enterprises are highly complex systems which

consist of multiple business and process units geo-

graphically distributed and also with different scales

working together. The simultaneous inclusion of eco-

nomic and environmental concerns within enterprise

structures requires high ﬂexible modeling systems,

capable of integrating the different scales and levels

of the organization. The integration of operations and

environmentaldata is one of the problems faced when

developing environmental decision support systems.

This paper proposes the reuse of an existing se-

mantic model in order to consider the addition of the

environmental system representation within the dif-

ferent supply chain decision levels, namely the life

cycle assessment. In order to carry out such a data

integration as well as to understand, analyze, syn-

chronize and ultimately improve the design and oper-

ation of complex ﬂexible production systems we be-

lieve that a promising avenue is to represent the whole

system in an adequate ontological model, which cap-

tures the environmental and operational features rel-

evant for managers to support decision making pro-

cesses. Speciﬁcally, the integrated enterprise ontol-

ogy project by (Muñoz et al., 2011), which repre-

sents an integrated enterprise framework is reused.

Thus, this work exploits the link between transac-

tional and analytical systems and semantic and quan-

385

Muñoz E., Capón E., Laínez J., Espuña A. and Puigjaner L..

Ontological Framework for Integrating Environmental Issues within Sustainable Enterprise - Enhancing Enterprise Decision-making.

DOI: 10.5220/0004141103850388

In Proceedings of the International Conference on Knowledge Engineering and Ontology Development (KEOD-2012), pages 385-388

ISBN: 978-989-8565-30-3

 2012 SCITEPRESS (Science and Technology Publications, Lda.)

titative models to propose sustainable solutions for

the design and operation of supply chains.

On the application of knowledge representation to

environmental issues, (Kraines et al., 2006) present

a framework for expert knowledge sharing and dis-

covery for integrated environmental assessment of

technologies and processes associated with industrial

ecology. A Life Cycle Assessment Ontology has been

proposed by (Brascher et al., 2007), including the do-

main represented by the ISO 14040 resulting in an

ontology containing the four phases of the LCA.

2 ONTOLOGY FRAMEWORK

The proposed ontology supports different activities

by streamlining information and data integration by

means of an integrated model which captures the ac-

tivities developed along the different levels of the en-

terprise structure resulting in an integrated decision

making framework. The ontology provides the shared

and common domain structures required for the se-

mantic integration of information sources, resulting

in a competitive advantage.

Thus, the developed ontology adopts as base

model a previous enterprise conceptualization pre-

sented by (Muñoz et al., 2011) which supports and

integrates the different hierarchical enterprise levels.

The reuse of this model allows an automatic incorpo-

ration of the previous domain, providing the shared

and common structures required for the semantic in-

tegration of information sources. What is more, when

a previous model is reused, there is a chance of im-

provement, since a new revision of the area is carried

out and a new part of a speciﬁc area or a new spec-

iﬁcation is included. In this paper, additional model

classes, properties and axioms have been developed

to include the environmental functionalities. As a re-

sult, the environmental assessment model is described

and integrated with the enterprise model. Extended

reference of the integrated enterprise ontology can be

found in the original paper by (Muñoz et al., 2011).

2.1 Domain Deﬁnition

On the one hand, the domain of the ontology com-

prises the whole enterprise entity, including those ac-

tivities related to the operational, tactical and strate-

gic levels. Speciﬁcally, the conceptualization and

management of operational concepts (physical mod-

els, procedures, functions and processes) is based on

ANSI/ISA-88 and ANSI/ISA-95 (for Measurement

and Control, 2007) process standard, which catego-

rize and examine the relationships among them. On

the other hand, the environmental domain is rep-

resented based on the life-cycle assessment (LCA)

methodology standardized in the ISO 1404X series

(ISO14001, 2004) for setting an environmental man-

agement system (EMS).

2.2 Methodology

The methodology adopted in this work regarding

the ontology development consists of the continuous

improvement cycle using PDCA (Plan, Do(Study),

Check and Act) and replanning phases, as presented

in (Muñoz et al., 2011). This methodology encom-

passes the steps proposed by two existing methodolo-

gies, namely "Methontology" (López et al., 1999) and

“On-To-Knowledge” (Sure and Studer, 2002). The

former methodology supports the entire life-cycle of

ontology development, whereas the latter allows to

present knowledge efﬁciently and effectively by the

analysis of usage scenarios. Since the new ontologi-

cal model is based on a previous one (Muñoz et al.,

2011), the result of the re-planning and revisiting of

the different phases in the PDCA cycle is presented

next.

Plan Phase. This stage captures the new require-

ments and speciﬁcations regarding the environ-

mental management domain, and next adequately

documenting them. The possible knowledge

sources are deﬁned for expanding the original do-

main. Basically, the environmental domain re-

lies on deﬁnitions used by internationally recog-

nized organizations, environmental agencies and

research reports from the literature (Garrido and

Requena, 2011; Bojarski, 2010).

Do Phase. At this stage, the principal components

of the conceptualization model related to the en-

vironmental domain are established. Glossary

of terms, concepts and properties, hierarchies,

the taxonomy, class and instant attributes are de-

scribed. Thus, the equivalence between the terms

in the environmental assessment model and the

enterprise ontology model must be established.

The formalization of all the contents should agree

with the knowledge sources. Next, the model of

the ontology is formalized considering the previ-

ous information.

In this case, the OWL ontology editors used have

been Protégé as main editor and Swoop as com-

plementary editor, being both freeware and also

robust softwares.

Check Phase. In this stage, the language and the

conceptuality are checked in order to standard-

ize them with the support of expertise and ex-

KEOD2012-InternationalConferenceonKnowledgeEngineeringandOntologyDevelopment

386

perts. The reasoning of the ontological model is

done, which is one of the most important tasks.

Additionally, a short informatics application has

been developed in order to test the ontology in the

main application environment. Common defects

include inconsistent ontologies and unsatisﬁable

concepts. In this work the RacerPro reasoner from

Protégé and Pallet reasoner from Swoop were

used as reasoners for testing the ontology.

Act Phase. Having found defects in the ontology,

their resolution can be non-trivial, requiring an

exploration of remedies with a cost/beneﬁt anal-

ysis. In this case, one would like to generate re-

pair solutions that impact the ontology minimally.

Particular care and effort must be taken to ensure

that ontology repair is carried out efﬁciently. The

necessary documentation of the implementation is

fulﬁlled for the maintenance ontology task.

2.3 Models Usability

In order to exploit the full potential of the ontologi-

cal model for connecting transactional and analytical

systems, Java has been used as a high-level program-

ming language, because it provides a good versatility,

efﬁciency and security. The code was built using the

platform NetBeans IDE 7.0.

The application of the ontological model for the

environmental performance assessment takes place

within the different enterprise decision layers, which

are integrated by means of the enterprise ontology

project. In this particular work, decision-making con-

cerns the operational and strategic levels.

The operation of the framework consists of re-

trieving the environmental mid-point categories re-

lated to the functional units and elements in the LCA

assessment by querying the environmental database

using the Java code, the values for the ontologies in-

stances that require environmental assessment. Once

such values are obtained, they are stored in the on-

tology. Next, when the optimization model needs the

data for formulating the problem, such information is

retrieved from the ontology as well.

The dimension of the ﬁnal ontological model con-

sists of 276 classes, 67 restrictions, and 224 object

properties stemming from the enterprise model itself,

and 108 classes, 34 restrictions, and 74 object prop-

erties characterizing the environmental management

domain. These components make the ontology rea-

soning and its use possible. The reasoning time for

the consistency of the model and classes is 1.062

and 0.281 CPU s respectively in an Intel-Core 2 at

2.83GHz in a successful compilation.

The analytical systems for taking decisions are

based on mathematical optimization, speciﬁcally the

centralized approach to supply chain design and plan-

ning presented by (Laínez et al., 2009) and the STN-

formulation (Maravelias and Grossmann, 2003) to

production scheduling are considered in this work.

3 CASE STUDIES

The case studies are based on a supply chain network

design-planning problem presented by (Laínez et al.,

2009). It consists of three suppliers, four potential

locations for the processing sites and the distribution

centers in a planning horizon of ﬁve annual periods.

The production process fulﬁlls the demand of six mar-

kets that entails two ﬁnal products and one intermedi-

ate product.

The strategic analytical optimization model pre-

sented by (Laínez et al., 2009) also includes an LCA

formulation that computes the mid-point and end-

point impact categories based on the production and

distribution activities that are performed to optimally

satisfy the demand. Qualitatively speaking, the prob-

lem representation in the proposed ontological frame-

work results in 761 instances. The reasoning time for

the problem instances is 0.688 CPU s in a successful

compilation.

It is important to mention that each possible site

is fully represented in the ontology. Each produc-

tion plant (site) may contain a set of four equipment

technologies as presented by (Kondili et al., 1993), a

benchmark problem for the scheduling of batch pro-

cess industries. The production process consist of ﬁve

production tasks and nine states, namely three raw

materials, two ﬁnal products and four intermediates.

Speciﬁcally, each site is described by 111 instances,

which may be adequately used to take operational de-

cisions. Precisely, the scheduling optimization model

has been adapted to also consider decisions related to

environmental considerations of the plant.

The analytical optimization model must be pro-

vided with the necessary information, which is de-

rived from the ontological model and the related data

contained in the database. Additionally, the ontologi-

cal model optimizes the way in which the databases

are distributed along the enterprise structure. As a

result, databases are well located and their data are

easily available and can be transformed into valuable

information.

In order to generate the required inputs for the

optimization model which has been implemented in

GAMS and solved using the MILP solver CPLEX 9.0,

the Java application is used. Such code generates the

.txt ﬁles which are called by the optimization prob-

OntologicalFrameworkforIntegratingEnvironmentalIssueswithinSustainableEnterprise-EnhancingEnterprise

Decision-making

387

lem.

The results of the optimization model are identical

to those reported in the original paper. Furthermore,

the previous results can be dated back to the ontolog-

ical model for further exploitation by the other de-

cision levels, such as the operational system of each

site. This can be achieved by automatically updating

the databases with the resulting optimization data.

4 CONCLUSIONS

This ontology enhances the way for achieving a suc-

cessful enterprise decision making supporting tool

which adapts and recognizes the different elements

found through the hierarchy models that are associ-

ated with the whole supply chain and allows to assess

the environmental performance of the enterprises by

communicating to environmental life cycle databases

and to analytical models.

Moreover, a general semantic framework is pro-

posed, which is able to model any enterprise partic-

ular case and its environmental implications, prov-

ing its re-usability. Furthermore, it has been proved

the ontology usability by its application to an opti-

mization framework. As a whole, the main contri-

butions of this framework and the model behind are

re-usability, usability, higher efﬁciency in communi-

cation and coordination procedures within the enter-

prise in order to assess its environmental issues.

This work represents a step forward to support

the integration, not just communication, of different

software tools applicable to the management and ex-

ploitation of plant database information, resulting into

an enhancement of the entire process management

structure for aiding the automatic design and opera-

tion of more sustainable enterprises.

Further work is underway to unveil the full poten-

tial to implement a large-scale semantic web approach

to support business processes decisions in a sustain-

able enterprise.

ACKNOWLEDGEMENTS

Financial support received from CIMAT México is

fully acknowledged. In addition, ﬁnancial support

provided by research Project EHMAN (DPI2009-

09386) funded by the European Union (European Re-

gional Development Fund ERDF) and the Spanish

"Ministerio de Ciencia e Innovación" is fully appre-

ciated.

REFERENCES

Bojarski, A. (2010). Life cycle thinking and general mod-

elling contribution to chemical process sustainable

design and operation. PhD thesis, Universitat Politèc-

nica de Catalunya.

Brascher, M., Monteiro, F., and Silva, A. (2007). Life cycle

assessment ontology. In ISKO, editor, Proceedings

of the 8th Conference of the International Society for

Knowledge Organization, pages 169 – 177.

for Measurement, I. S. and Control (2007). Isa-88/95 tech-

nical report: Using isa-88 and isa-95 together. Tech-

nical report, ISA The Instrumentation, Systems, and

Automation Society.

Garrido, J. and Requena, I. (2011). Proposal of ontology

for environmental impact assessment: An application

with knowledge mobilization. Expert Systems with

Applications, 38:2462 – 2472.

Grossmann, I. E. (2004). Challenges in the new millen-

nium: product discovery and design, enterprise and

supply chain optimization, global life cycle assess-

ment. Computers & Chemical Engineering, 29(1):29–

39.

ISO14001 (2004). Environmental management systems -

requirements with guidance for use. Technical report,

ISO.

Kondili, E., Pantelides, C., and Sargent, R. (1993). A

general algorithm for short-term scheduling of batch

operations–i. milp formulation. Computers & Chemi-

cal Engineering, 17(2):211 – 227.

Kraines, S., Batres, R., Kemper, B., Michihisa, K., and

Wolowski, V. (2006). Internet-based integrated en-

vironmental assessment, part ii: Semantic searching

based on ontologies and agent systems for knowledge

discovery. Journal of Industrial Ecology, 10:37 – 60.

Laínez, J. M., Kopanos, G., Espuña, A., and Puigjaner, L.

(2009). Flexible design-planning of supply chain net-

works. AIChE Journal, 55(7):1736–1753.

López, M. F., Gómez-Pérez, A., Sierra, J. P., and Sierra,

A. P. (1999). Building a chemical ontology using

methontology and the ontology design environment.

IEEE Intelligent Systems, 14(1):37–46.

Maravelias, C. T. and Grossmann, I. E. (2003). New

general continuous-time state-task network formula-

tion for short-term scheduling of multipurpose batch

plants. Industrial & Engineering Chemistry Research,

42:3056–3074.

Muñoz, E., Capon-Garcia, E., Lainez, J., Espuña, A., and

Puigjaner, L. (2011). Ontological framework for the

enterprise from a process perspective. In SciTePress,

editor, Proceedings of the International Conference

on Knowledge Engineering and Ontology Develop-

ment, pages 538 – 546.

Sure, Y. and Studer, R. (2002). On-to-knowledge method-

ology. Technical report, University of Karlsruhe.

KEOD2012-InternationalConferenceonKnowledgeEngineeringandOntologyDevelopment

388