Improving operational risk management systems by formalizing the basel ii regulation with goal models and the isoiec 15504 approach

Improving Operational Risk Management Systems
by Formalizing the Basel II Regulation with
Goal Models and the ISO/IEC 15504 Approach
André Rifaut1 and Christophe Feltus1
1
Centre de Recherche Public Henri Tudor, 29, Avenue John F.Kennedy,
L-1855 Luxembourg-Kirchberg, Luxembourg
{Andre.Rifaut, Christophe.Feltus}@tudor.lu
http://www.tudor.lu
Abstract. The bankruptcy of financial institutions shows the rapid changes in
the risks profiles of financial systems and processes. Although financial institu-
tions have always managed the operational risks, the profile of this kind of risks
is changing due to the increasing international competitive pressure and the
evolution of the financial institutions’ operational systems relying more and
more on IT systems. This paper reports the results of the joint research with the
CSSF [1] focusing on the formalization of both the Basel II Accord and com-
pliant operational risk management (ORM) systems implementations. This for-
malization uses concepts of the ISO/IEC 15504 process assessment standard
and the concepts of strategy and policy. This structure of the model ensures the
traceability between the Basel II Accord and compliant ORM systems imple-
mentations, improves the formal validation of those systems and is more ade-
quate to represent all organizational levels of financial institutions.
1 Introduction
In Luxemburg, the stability of the financial system is at the core of the economic
stability of the country. The CSSF [1], which is the official authority for financial
institutions supervision, has the responsibility to define financial regulations and en-
sure their fulfillment. This task is not easy because more and more international regu-
lations are introduced, such as the IFRS [2], Sarbanes-Oxley Act (SoX) [2] and the
Basel II Accord [3]. Audit managers, risk managers (including security managers),
and compliance managers have developed standards addressing those regulations. For
instance, Coso [2], CobIT [2], ITIL [18] and ERM [2] are governance and risk man-
agement standards. However, up to now there is nearly no integration between the
regulations themselves and also between those standards. A joint research with the
CSSF aims at defining a method for ensuring a correct implementation of financial
systems compliant to Basel II regulation. The results [20,5] are based on quality
methods and techniques, mainly goal-based models and analyses used in goal-
oriented requirements engineering (GORE) [4]. The originality of the work lies in the
formalization of the Basel II Accord and Operational Risk Management (ORM) sys-
tems by using concepts of the ISO/IEC 15504 process assessment standard [6] and
the concepts of strategy and policy. This gives an adequate structure of the models at
Page 1 of 7

all organizational levels of financial institutions, ensures the formal traceability be-
tween the Basel II Accord and ORM systems, and improves their formal validation.
This paper summarizes and extends the results of the joint research with the CSSF,
focusing on the formalization of both Basel II Accord and compliant (ORM) systems
implementations. More information on the research results, the ISO/IEC 15504 stan-
dard, the Basel II Accord, and other standards such as ITIL are freely available on the
CSSF website [1]. The next section presents the main goals of this research and the
preliminary results. Section 3 shows the technique that has been created in the context
of the real case study concerning the Basel II Accord regulation and its implementa-
tion in financial institution. The last section summarizes the main results of this pro-
ject and presents the future works that will be done within the follow-up research
projects.
2 The Implementation of ORM Systems compliant to Basel II.
The Basel Committee has defined the operational risk as follows: it is the risk of loss
resulting from inadequate or failed internal processes, people and systems or from
external events. (§644 in [1]). As such, the operational risk encompasses all risks oc-
curring at the operational and technical levels (see Fig. 1), in particular, all risks of
the IT Software Engineering Processes (risks that concern project management, re-
quirements analysis, design, security, ...). The methods used in IT Software Engineer-
ing (e.g. for safety and security analyses) do not cover the analysis of this very broad
scope of risks.
The need for practical techniques is critical in order to help business units’ man-
ager to efficiently implement the core business processes that are under their respon-
sibility. Indeed, not only the Basel II Accord is imposing constraints on those core
financial processes, but also the other regulations (e.g. SoX, IFRS) are interfering on
the same processes. Moreover, each regulation stresses the importance on a different
but inter-related aspect. For instance, SoX stresses the importance on the reporting
system also concerned by the ORM of Basel II Accord. In addition to that decisions
about ORM system implementation must be made at the strategic, tactical, opera-
tional and technical levels. This increases the complexity of modeling and implement-
ing ORM systems, taking into account also that operational risks exist in every busi-
ness processes implying their strong relationships with new ORM systems. Last but
not least, those regulations are hard to understand due to their lack of structure and
lack of completeness. For instance, in the Basel II Accord there is no definition of
important concepts such as “ORM system”, “loss”, “loss event”, “expected loss”,
“unexpected loss”, ...
Page 2 of 7

Requirements engineering and goal-oriented methods. The GORE methods can
overcome the difficulties presented in the preceding section by formalizing the Basel
II Accord and the implementation of ORM systems. These methods can be used to
analyze and model systems at all organizational level, from Business Models up to
architectures [4]. Goal-oriented modeling languages are appropriate for that broad
range of models and they support formal analyses. However, in the case of the ORM
system, it is difficult to manage all of those large models and complex analyses.
Moreover, for validation purposes, it is important to refer to the concepts used in
organizations, such as strategic objectives, strategies and plans, key indicators, poli-
cies, SLAs, ... Within the Basel II Accord context, additional structuring mechanisms
has to be created on top of the usual goal-oriented concepts.
Technicallevel
Operationallevel
Tacticallevel
Strategiclevel
E.g. in the IT domain:
applications, components, …Procedures
Business
processes
Business
value
Objectives
Indicators
StrategiesPolicies
Goals
Indicators
StrategiesPolicies
Goals
Indicators
StrategiesPolicies
Standard view of
organizational layers
Concepts
and artifacts of
Goals
Indicators
StrategiesPolicies
Goals
Indicators
StrategiesPolicies
Goals
Indicators
StrategiesPolicies
Goals
Indicators
StrategiesPolicies
Technicallevel
Operationallevel
Tacticallevel
Strategiclevel
Business
processes
Business
value
Technicallevel
Operationallevel
Tacticallevel
Strategiclevel
Technicallevel
Operationallevel
Tacticallevel
Strategiclevel
Business
processes
Business
value
Business
processes
Business
value
Objectives
Indicators
StrategiesPolicies
Goals
Indicators
StrategiesPolicies
Goals
Indicators
StrategiesPolicies
Standard view of
Concepts
and artifacts of
Goals
Indicators
StrategiesPolicies
Goals
Indicators
StrategiesPolicies
Goals
Indicators
StrategiesPolicies
Goals
Indicators
StrategiesPolicies
Fig. 1. The pyramid is used in management methods (e.g. [7]). The lowest 4 artifacts are de-
fined with GORE models [4].
3 Formalizing Basel II and ORM with goal models and the
ISO/IEC 15504 approach
The general framework given in the Figure 1, represented by the pyramid, is a stan-
dard view of the organization [7] used in financial institutions (and other institutions).
The four organizational layers [8] – strategic, tactical, operational and technical levels
– use concepts adapted to handle decisions at their corresponding abstraction level –
that are mainly (respectively) business value [9], business processes, procedures and
technical artifacts ([10]) such as IT applications in the IT domain.
Page 3 of 7

ISO/IEC 15504 process assessment model. A first part of the structure is given by
separating the description of the core activities of the business processes from the
activities related to the capabilities of the business process (e.g. planning, work
product control, process documentation, performance measurement, performance
improvement, ...). As explained in [5], the benefits of this separation of concerns has
proven to be very useful during the verification and the validation of the goal models.
When describing process models with the ISO/IEC 15504 standard this separation of
concerns is imposed. This new standard has been designed to be applicable for any
business processes and is no longer limited to software engineering processes [20].
Objectives, strategies, policies and indicators. Those concepts (bottom of Figure 1)
detail complementary aspects needed for designing business processes, procedures
and technical artifacts. They are similar to organizational concepts needed in order to
structure and formalize the links between each of the organizational levels [11].
When designing lower-level artifacts from higher level ones and when verifying
the link between two organizational levels, one has to distinguish between the main
objectives to be fulfilled, the strategy describing the approach to fulfill these objec-
tives, the roles and responsibilities (policies) of the resources that will implement the
strategies. Indicators are defined when there is a need for some monitoring, control,
supervision or measurement concerning objectives, strategies or policies. Strategies
and policies must be complementary and consistent with each other and they must
fulfill the objectives.
Page 4 of 7

Fig. 2. Basel II ORM (left side) partially implemented (right side). Upside-down arrows shows
that the implementation contributes to the each level of ORM.
The formal definition of those 4 concepts uses goal-oriented techniques [4,5]. For
the indicators, our work is based on the Goal-Question-Metric method (GQM) [12].
Policies give a description of the roles and responsibilities (in accordance with [13]
and policy management [14,15,16]) and allow detailing the authorizations, obliga-
tions (and their delegations), accountabilities, and separations of duties. Strategies
give a description of the main approach or steps to fulfill given objectives. Our work
follows [17] where strategies are integrated with goal-oriented analysis. For the sake
of separation of concerns, responsibilities (and related aspects) are not defined in
strategies but only in policies. Note that in financial institutions, the description of
policies recalls its related objectives and strategies. This is also sometimes the case of
strategies that gives a short description of their corresponding policies (i.e. descrip-
tion of roles and responsibilities). However, it is found essential to separate those
descriptions when designing and analyzing those policies and strategies.
For instance, in the Figure 2, the diagram shows the model of the strategic level
(topmost) and operational level (bottom). Only objectives are shown for those two
levels. In between, at the tactical level, the objectives and indicators of business proc-
esses are shown. The left part of the diagram shows the Basel II Accord formalization
of ORM. The right part presents a partial ORM system implementation using ITIL
[18]. The links between the two models are formally analyzed [5].
Page 5 of 7

4 Conclusions and Future Works
Building upon a method that has been defined within the setting of a real-case
study in financial institutions, the Basel II Accord, new results are presented in this
paper aiming at giving a simple but integrated set of concepts – goals, indicators,
policies and strategies – which can be used to design financial systems compliant to
regulations and structure their analysis in relationship with the artifacts commonly
used in financial institutions – business models, business processes models, proce-
dures and more technical artifacts. The formalization of goals, indicators, policies and
strategies independently from each other allows analyzing and recording the design
decisions across all organizational levels, making easier the link with the regulation.
The main advantage of this method is that it keeps the structuring power of the
ISO/IEC 15504 capability model that can be used to discover weaknesses and opera-
tional risks in the business process implementation with the method explained in [19].
Based on the same techniques as in [15], a prototype implementation is under devel-
opment.
The current and future works of the authors focus on a constructive method aiming
at giving an effective support for financial business process design (compliant to
regulations), establishment, assessment, improvement, governance and benchmarking
[5]. In particular, a risk and value analysis method is under development adapted to
process assessment, improvement and governance. Some support is also given to an-
other research made by experts in DPM [21]. The aim of those experts is to ground
digital policy management in sound non-federated distributed IT systems that en-
forces policies fulfillment even outside the traditional IS frontier of each institution.
Finally, the current project with the CSSF is still in progress with results that are ex-
tended to the IFRS [2] concerning the management of unquoted assets (IFRS-IAS39)
[2]. In addition to model this regulation and the systems compliant to it, the relation-
ship between IFRS-IAS 39 and Basel II can be analyzed and alternative compliant
implementations of integrated systems can also be designed.
References
1. CSSF: Commission de Surveillance du Secteur Financier. The firsts results of the joint pro-
ject are freely downloadable at http://www.cssf.lu/index.php?id=130 .
2. IFRS: International Financial Reporting Standards, IASCF, USA. SoX: Sarbanes Oxley Act
of 2002, USA. COSO: Internal Control – Integrated Framework, CSOTC, USA. CobiT®:
Control Objectives for Information and related Technology, ISACA, USA. ERM: Enterprise
Risk Management – Integrated Framework, CSOTC, USA.
3. Basel Committee on Banking Supervision, “International Convergence of Capital Measure-
ment and Capital Standards”; BIS; Basel, June 2004.
4. A. van Lamsweerde, "Goal-Oriented Requirements Engineering: A Guided Tour". Invited
minitutorial, Proc. RE'01 - International Joint Conference on Requirements Engineering,
Toronto, IEEE, August 2001, pp.249-263.
5. André Rifaut, “Goal-Driven Requirements Engineering for Supporting the ISO 15504 As-
sessment Process”, EuroSPI 2005, Budapest
6. ISO/IEC 15504, “Information Technology – Process assessment”, (parts 1-5), 2003-2006
(see website [1] for details about this standard).
Page 6 of 7

7. Anthony, R. N. Planning and Control Systems: A Framework for Analysis. Harward Univer-
sity, Boston, USA, 1965
8. Henderson, J. and Venkatraman, N., “Strategic alignment: Leveraging technology for trans-
forming organizations”. IBM Systems Journal, 1999, 38.
9. Osterwalder and Pigneur. An Ontology for e-business models. In “Value Creation from E-
Business Models”, Wendy Currie ed., Butterworth-Heinenmann. Apr 2005.
10. Robson W, Strategic Management and Information Systems, Pitman, 1997.
11. Chaffey et al. (2005) - Business Information Systems: Technology, Development and Man-
agement for the E-business, Prentice Hall.
12. Van Solingen, “The Goal/Question/Metric Method: A Practical Guide For Quality Im-
provement of Software Development”, McGraw-Hill, Jan. 1999
13. René Wies, “Using a Classification of Management Policies for Policy Specification and
Policy Transformation”. In Proc. ISINM '95, Santa Barbara, California, May 1995.
14. N. Damianou, N. Dulay, E. Lupu, and M. Sloman, “The ponder policy specification lan-
guage” In Morris Sloman, (ed), Proc. of Policy Worshop, 2001, Bristol UK, January 2001.
15. A. Schaad and J. Moffett. “Delegation of obligations.” In IEEE Policy Workshop, 2002
16. Qingfeng He and Annie I. Antón, “Deriving Access Control Policies from Requirements
Specifications and Database Designs”, ICSE 2005.
17. Rolland C., N. Prakash, A. Benjamen, “A multi-model view of process modelling.” Re-
quirements Engineering Journal, p. 169-187,1999.
18. ITIL: IT Infrastructure Library – Service Support, Service Delivery, published by OGC,
London. (see website [1] for details about this standard).
19. A. Rifaut, M. Picard and B. Di Renzo, “ISO/IEC 15504 Process Improvement to Support
Basel II Compliance of Operational Risk Management in Financial Institu-
tions”,International Conference SPiCE 2006.
20. B. Di Renzo, M. Hillairet, M. Picard, A. Rifaut, C. Bernard, D. Hagen, P. Maar, D. Re-
inard, “Operational Risk management in Financial Institutions: Process Assessment in Con-
cordance with Basel II”, International Conference SPiCE 2005.
21. J.-H. Morin and M. Pawlak, “Towards a Global Framework for Corporate and Enterprise
Digital Policy Management”, SoftWars conference, Las Vegas, USA, Dec 11, 2005.
Page 7 of 7

Improving operational risk management systems by formalizing the basel ii regulation with goal models and the isoiec 15504 approach

More Related Content

What's hot (7)

Viewers also liked (20)

Similar to Improving operational risk management systems by formalizing the basel ii regulation with goal models and the isoiec 15504 approach (20)

More from Luxembourg Institute of Science and Technology (20)

Recently uploaded (20)

Improving operational risk management systems by formalizing the basel ii regulation with goal models and the isoiec 15504 approach