AD ALTA
JOURNAL OF INTERDISCIPLINARY RESEARCH
in a public university are stored in the data warehouse via a
developed web-based application.
Demonstration of the descriptive and the predictive
decision making possibilities of the proposed solution via
online analytical processing and machine learning practical
examples respectively.
Utilization of the estimated maximum recovery time to
compute the availability rates of the same safety – critical
operations.
Based on the above stated objectives the rest of the paper is
organized as follows:
Section 2 is devoted to the problem statement and the provision
of the necessary background information. Section 3 includes a
brief delineation of the business continuity recovery time effort
estimation mathematical approach entitled Business Continuity
Points for which a data warehouse schema is proposed as a
database solution to host data for safety critical business
functions.
Business intelligence and data warehouse
fundamentals are also included. Section 4 is used for analysis of
the conducted results, including a thorough discussion with
respect to the accuracy of the proposed solution in business
continuity and safety management operations. The article is
finalized with the conclusions and the future research directions.
2. Problem Statement and background information
“A Safety Critical System is such a system which has the
potential and may cause accidents either directly or indirectly.
Failure of such systems can result in loss of life,
property
damage, environmental harm and financial loss. Safety is
dependent on proper operations of such systems” (Srinivas
Acharyulu & Seetharamaiah, 2015). It is thus, important to
classify such systems as highly critical in terms of recovery
priority establishment stemming from the computation of their
recovery time, and bearing in mind that such systems should
operate without or with minor interruptions. As a consequence,
the incorporation of mathematical tools and software solutions
for criticality ranking of industrial business functions which are
dependent on safety critical
systems becomes a clear necessity.
An interesting mathematical approach by Torabi et al (2014)
refer to proactive recovery time estimation of critical business
functions based on multiple criteria decision making. However,
the method implements criticality ranking for a group of BFs and
does not focus on the peculiarities and the unique technological,
user and process related features as well as the environmental
parameters of an individual BF.
Moreover, software tools which have been designed and
developed for the BCM domain (Šimonová, S., & Šprync, O.,
2011) though proactive, they serve as tools which manage
operational failures by focusing exclusively on the technical
aspects of the business functions, and do not take into account
the environmental aspects (i.e. experience of the end user, users’
motivation) of an individual business function. Additionally,
mathematical models which are proposed in combination with
ICT - based solutions (Sahebjamnia et al, 2014) behave as
reactive (not proactive) BCM and disaster recovery planning
(DRP) solutions for the resumption of critical operations after
their failure.
The current contribution is proposed based on the gap which is
realized from the study of the available literature, according to
which none of the sophisticated BCM tools and methods
computes the recovery time of individual business functions
based on input data that stem from the unique technological and
the environmental features of this function. Additionally, “data
collection is an important activity throughout the BCM
development process” (Engemann & Henderson, 2012) and
every “resilient organization, through an enhanced sensing
capability, integrates business intelligence in order to improve
situational awareness” (Starr, 2003). The current research is
devoted to the construction of a business intelligence software
tool which can efficiently support data collection towards the
precise classification and recovery time estimation of a given
business function. Moreover, the computed timeframes can be
used as a standard input for system availability measurement for
safety – related functions in public organizations. The data
warehouse features are conceptualized based on the business
continuity points method. In the current work, real data from a
public university are used for validating the initial BCPTs
method, the suggested BI tool as well as the availability result
for safety functions in the public institution.
3. Tools and Methods
3.1 The Business Continuity Points (BCPTs) approach
The approach (Podaras et al, 2016) focuses on the proactive
estimation of the recovery time effort for an individual business
function and its corresponding criticality ranking. The
algorithmic process for calculating the Recovery Time is below
depicted (Fig.1).
Fig. 1. The summarized model of the BCPTs Approach.
(Source: own work).
For the better interpretation of the derived results we briefly
mention that for the estimation of the recovery time effort we
have to consider a set of recovery complexity parameters.
3.1 Unadjusted Business Function Recovery Points (UBFRP)
In order to compute the specific value human and application
level actors have to be considered along with their corresponding
impact (weight) on the recovery process. Moreover, the number
of the involved processes and the level of complexity of each
process has to be calculated. Summing up all these unadjusted
parameter values, the unadjusted business function recovery
points variable is computed (Fig.1).
The general function which is utilized to compute the UBFRP
value is provided by Eq. (1):
1
1
1
=
(
*
)(
*
)(
*
)
i
i
i
n
n
n
i
HA
i
AP
i
BF
i
i
i
UBFRP
HAW
APW
BFW
=
=
=
+
+
∑
∑
∑
(1)
Where, HA=Human Actor i, AP=Application or Technical Actor
i, BF= Business Function (or process or activity) i and W is the
weight or importance of the given parameter. The corresponding
values for each level of importance (W) are defined as follows:
Simple: 0.5, Average: 1 and Complex: 1.5.
Example: based on the available BCM data from the Columbus
Public College, the following parameters are considered,
regarding the safety critical operation named as emergency
communication:
Human Actors: 6 Human Actors (including 1 Process Manager
(Complex Level: 1.5), and 1 backup employee (Average Level:
1)
Technical Actors (mainly software tools): 5 defined Technical
Actors (SW and IT infrastructure) , including 2 complex (1.5), 2
average (1) and 1 simple (0.5).
8 delineated highly important processes (complex: 1.5) based on
the function description and the recovery strategy overview.
Thus the UBFRP value is computed as follows:
- 358 -