Risk analysis is a process that involves mitigation of errors during various processes of software development. Early identification of risks should shorten the software development completion time. It will also provide efficient solutions. A complete risk analysis along with identification, monitoring and management plan is sure to ensure high quality software products. In this paper reviewed different perception of reusability and the techniques followed to mitigate the risk arose while reusing components.
Keywords |
| Software Engineering, Software Reusability, risk Management, Object oriented programming |
INTRODUCTION |
| Software Engineering is a systematic approach to development, operation, maintenance and up gradation of software.
In general, Software Engineering is a discipline which provides tools and techniques to develop quality software in an
orderly fashion. It provides an engineering approach that is used to build software for end users using well defined, well
managed, consistent and cost effective processes.The Software Development Life Cycle (SDLC) is the sequence of
different activities that take place during the development. Any software development consists of five phases namely
Requirement Analysis, Design, Implementation, Testing and Maintenance (Pressman 2000). The SDLC begins with the
identification of the requirements of a software and ends with the formal justification for the development of software
against that requirement. SDLC is the period of time that starts, when a software product is conceived and ends when
the product becomes dormant. |
NEED FOR RISK ANALYSIS IN SOFTWARE PROJECT |
| Occurrences of errors during the execution of a program need to be predicted before implementation. This is
generally done through Risk analysis. For instance, classifying data types with variables would reduce the risk of
errors. Literatures have pointed out the requirements of risk analysis for these types of problems.A risk is a potential
problem which may or may not occur. Though software development is a very difficult process that determines risks
and taking practical measures to avoid them, risk analysis becomes mandatory. In addition, to manage the risks, in
order to reduce the complexity and the efforts taken for it, also becomes mandatory (Boehm 1991). |
| A.)Software Risks: The risk has two characteristics, |
| 1) Uncertainty: The risk may or may not occur. |
| 2) Loss: Unwanted consequences or losses will occur if risk is present. |
| To quantify the levels of uncertainty and the degree of loss associated with the risk, the type of risks should be
categorized as follows: Risks which are external to program coding have been delimited to the study for this
research. Four of these risks namely |
| 1) ‘People Risk’ would lead to reduction of efficiency, |
| 2) ‘Project Risk’ is sensitive to project cost |
| 3) ‘Technical Risk’ is related to the time of implementation and |
| 4) ‘Business Risks’ influence the software viability. Hence these issues are considered for the research study. |
CONCEPT OF REUSABILITY |
| A system is said to be reusable, if it has the capability of using its components in an orderly way. In short, a
part or whole of one system can be incorporated to assemble a new system. As far as software is concerned, it makes
the reusability simpler and competent to frame a new module which is free from bugs and risks (William Frakes and
Carol Terry 1996). Table 1 depicts the merits and demerits involved in the concept of reusability. |
LITERATURE REVIEW |
| The technique of programming is widely brought up into two categories which are as follows:- |
| 1) Procedure oriented programming |
| 2) Object oriented programming |
| These two vital techniques ie., Procedure Oriented Program and Object Oriented Program are different from each
other (Firesmith 1993 and Ryder et al 2005). |
| The former language which is procedure oriented usually follows algorithm-centric approach in which a program is
driven by an algorithm that traces its path of execution from beginning to end. Data is an external entity the operation
of which is defined upon by the algorithm. Fundamentally, these classes of programming languages are characterized
by data that are considered and separated from the operations or program part. Algorithm which is the primary most
stepping stone, which being the driver, with data being subsidiary to the algorithm, the procedure oriented program
just breaks up a larger job into a number of subtasks and subprograms independently and conveniently as functions
and subroutines. The functions and the subroutines which makes a complex job with best output, put together as a
whole forms an efficient, easily debugging capability rich and a best programming part. |
| The main notion of the procedure oriented language is to make the process of debugging a program and to increase
the number of reusable procedures in programs. The importance of program reuse is thus realized by software
professionals. By adopting reusability techniques in programming, the cost of building software would be reduced
considerably which have paved way for the emergence of Object Oriented Language. Khan Al-A’ali and Girgis (1995)
have described the Object oriented programming language to be data or Object centric. |
| These programming languages which are described to be object oriented plays role in the promotion of objects
which in turn helps in identification of the relationship between data and object that enhances creativity in the
technique of reusability. When certain common data and objects are used in various different programs, it will greatly
contribute in the reduction of time of development for the programmers and thereby increasing the efficiency of the
program. Reusable components are chosen in such a way that they will be best suited for the activities in the
implementation phase, to reduce risk. The implementation of code reusable components is mainly focused in this
thesis. |
| Nunamaker et al (1989) put forth brief ideas of approach in order to understand the highly complex process of
development of software and to improve the productivity of software including the using and designing of automated
software development tools. The study of human factors in software development and application of software
productivity measurement and evaluation technique are some other processes. |
| An environment to facilitate the reuse of software components is described. Such an environment supports the
identification, creation, categorization, storage, selection, usage, and maintenance of reusable components. A meta
system environment that allow users to define functionalities, structures, and constraints of various software
components is discussed. Informations regarding all these components is being used by a knowledge-based system in
order to support the selection, configuration, and distribution of the components that are reusable. |
| Ramamoorthy et al (1993) have stated that the implementation of risk management over the reusability technique of
software development is based on two knowledge tools, |
| A.) Risk assessment |
| This tool will greatly help the knowledge based systems that are involved in analysing all the metric data to
high risk and low risk components. Influence Diagram Based Expect System (IDES) helps in the estimation of
expected loss of each risk item. The influence of problem and decision on it, rules and inference engine are solved by
the system. The conflict resolution strategy is also provided by it. |
| B.) Consistency management |
| Consistency management system ensures proper operation of the system after correction and measurement of
the relationship between the components in each phase. This approach is called Assumption–Based Truth
Maintenance System (ATMS) which is used by any team for dependency relationship between the software object and
data throughout the life cycle of the development of system. This tool can helps in the prevention of disaster and also
in the reduction of impacts of the risks. |
| Domain Analysis method is to determine similar software patterns from the experiences of development to
satisfy the current needs of customers. Past experience guides to reuse the existing software products and improves
the quality of the delivered product. The system analyst has to undertake a full fledged process of analysis of the
current problems of the customer and finding the similarity of requirements from any completed projected
requirement. Domains of products have to be identified for comparing the current problem domain to minimize the
efforts for developing the system. This technique of analysis of domain was declared by Basili et al (1994). |
| Object Oriented Metrics rely on programs and are estimated through an empirical study which focuses on
improvement process which increases the necessity of using software measures and metrics. The need for such metrics
was particularly notable when an organization was adopted for any further enhancement in programs. An automated
data collection tool was designed and implemented to collect an empirical sample of these metrics at two field sites to
demonstrate their feasibility and sorted out ways in which project managers may use these metrics for process of
improvement by the teams of Chidamber and Kemerer (1994). |
| Paul Raymond et al (1996) have carried out a elaborated discussion in their paper, the use of an object oriented
data modeling approach for quality and risk requirements pertaining to software projects. The proposed approach
provides an environment for users to develop individual views and permits the system to process semantically
heterogeneous queries. Here a temporal modeling framework is employed which allow the users to capture temporal
semantics of information associated with software metrics. Formalism for temporal modeling of metrics data using a
set of temporal relation is also created. Further, an object oriented query model to receive semantics associated with
software quality and risk is obtained for modeling temporal semantics of software data. To express queries, the notion
of scenario was also introduced where one scenario is identified through raw values or processed data. Temporal
modeling is used to express complex scenario spanning various metrics and to help in identifying risk and also for
evaluating the quality of the software project at various levels of abstraction. |
| The higher conceptual studies on examining the process how the organizations actually exploit the reuse
technologies and evaluates the reuse factors are reported. These factors finally affects the rate of reuse in an
organization thereby to attempt to enhance the measurement of the rate of reuse. |
| The effectiveness of reuse by establishing conceptual foundations in the literature for reuse and conducting an
empirical investigation of organizations using Ada technology is reported. This study established a wide difference
among the software reuse criteria such as domain, humans, tool, organization, software metrics, and environment. |
| The rate of reuse(as shown by Nam Yong Lee and Litecky (1997)) significantly depends upon |
| 1) reuse capability, |
| 2) software development effort, |
| 3) object-oriented design capability, |
| 4) repository development effort, |
| 5.) Ada technology capability, and |
| 6.) domain capability. |
| The management of changes in reusable software was discussed and carried out by David Kane et al (1997)
and his fellow members. Organization which undertake software reuse need to manage changes in component which
grow due to expanded usage. Often, this is an expanded life cycle of software asset. The six patterns to support reusable
software that emerged out of this discussion report are |
| 1) develop a shared platform |
| 2) maintain the reuse platform identity |
| 3) integrate reuse and tie to the bottom line |
| 4) reuse code |
| 5) treat reusable components like product and |
| 6) merge clones. |
| Each of this pattern must be described with their problem statement and with their solution of reusable
software.United State Navy Software Maintenance Organization management has created a process improvement
activity for upgrading the overall quality of internal software maintenance.This was exclusively explained by Charette
et al (1997). It provides a simple consensus method of risk assessment and mitigation. |
| The primary focus of this approach is on addressing the real risk rather than perceived ones. They state that
there is a significant difference between maintenance oriented risk management and development oriented risk
management. They have listed a suggestion of specific risk management information as input to specific risk
management process. |
CONCLUSION |
| In any software development set up, professionals differ from each other in their attitude, technical skill and
ability. Talented software professionals form a great asset to an organization. They, therefore reduce the levels of risks
involved in successful product development. |
| The extensive literature reviewed in this paper, has indicated that there is no model available for adaptive
reusability with reusable components in risk analysis for object oriented programs.An innovative model has been
proposed in the name of ‘Proficient Risk Extenuation Model (PRE-M) for Adaptive Software Reusable
Components’ and the design and developmental details along with its validity should be measured to focus on
mitigating the risk occur in software component reusability. |
| |
Tables at a glance |
 |
| Table 1 |
|
| |
References |
- Basili V.R., Bri L.C. and Thomas W.M, âÃâ¬ÃËDomain Analysis for the Reuse of Software Development ExperiencesâÃâ¬Ãâ¢, Experimental Software Engineering Group (ESEG), Vol. 11, pp. 86-95,1994.
- Boehm B. W., âÃâ¬ÃËSoftware Risk Management: Principles and PracticesâÃâ¬Ãâ¢, Journal of IEEE Software, Vol.8, No 1, pp. 32-41,1991.
- Charette R.N., Adam K.M, White M.B., âÃâ¬ÃË Managing risk in Software MaintenanceâÃâ¬Ãâ¢, IEEE Software ,Vol.14, No.3,pp.43-50,1997.
- Chidamber S.R. and Kemerer C.F., âÃâ¬ÃËA metrics suite for object oriented DesignâÃâ¬Ãâ¢, IEEE Transaction on Software Engineering, Vol. 20, No.6, pp. 476-493,1994.
- David Kane, âÃâ¬ÃËManaging Change to Reusable SoftwareâÃâ¬Ãâ¢, PLoP 97 Conference, pp.1-12,1997.
- Firesmith D.G., âÃâ¬ÃËTesting Object Oriented SoftwareâÃâ¬Ãâ¢, Proceedings of Eleventh International Conference on Technology of Object Oriented Languages and Systems, pp.406-427,1993.
- Khan Al-Aali and Girgis, âÃâ¬ÃËObject Oriented Programming for Structured Procedural ProgrammingâÃâ¬Ãâ¢, Journal of Systems and Software, Vol. 17, pp.111-117,1995.
- Nam Yong Lee and Litecky C.R., âÃâ¬ÃËAn Empirical Study of Software Reuse with Special Attention to AdaâÃâ¬Ãâ¢, IEEE Transaction on Software Engineering ,Vol.23, No.9, pp.537-549,1997.
- Nunamaker J.F. and Chen M.,âÃâ¬ÃËSoftware productivity: a framework of study and An approach to reusable componentsâÃâ¬Ã⢠System Sciences, Software Track, Proceedings of the Twenty-Second Annual Hawaii International Conference, Vol. 2, pp.959-968,1989.
- Paul Adams, Cornelia Bolyreff, David Nutter and Stephen Rank, âÃâ¬ÃËAdaptive Reuse of Libre Software Systems for Supporting Online CollaborationâÃâ¬Ãâ¢, Proceedings of the fifth workshop on Open source software engineering, pp.1-4,2005.
- Pressman R.S., âÃâ¬ÃËSoftware Engineering: A PractitionerâÃâ¬Ãâ¢s ApproachâÃâ¬Ãâ¢, Fifth Edition, McGraw-Hill International Edition, 2000.
- Ramamoorthy C.V., Chandra C., Ishihara S. and Ng Y., âÃâ¬ÃËKnowledge based tools for risk assessment in software development and reuseâÃâ¬Ãâ¢, Proceedings of the 1st International Conference on tools with Artificial Intelligence, pp.364-371,1993.
- Ryder B.G., Mary Lou Soffa and Margaret Burnett, âÃâ¬ÃËThe impact of software engineering research on modern programming languagesâÃâ¬Ãâ¢, ACM Transactions on Software Engineering and Methodology, Vol. 14, No. 4, pp. 432-477,2005.
- William Frakes and Carol Terry, âÃâ¬ÃËSoftware reuse: metrics and modelsâÃâ¬Ãâ¢, ACM Computing Surveys, Vol.28, No.2, pp 415-435,1996.
|
Menu