Abstract
The software development process involves wide range of variables that requires effective monitoring and control for the successful completion of the and projects. In order to ensure quality product development and delivery of services various initiatives to make the processes very efficient are proposed. CMMI frame work is the most accepted way to improve the process in order to make the operation very efficient.
The study undertaken here involves a detailed introspection into various steps of CMMI levels of maturity and hence to identify important areas that requires further attention. In addition managing risks of software projects is not a well attended area which requires inputs for successful completion of the projects. CMMI model incorporates necessary inputs to help the managers for identifying and mitigating the risks hidden in a software development project. The report contains a detailed survey on the CMMI framework being followed by software industries for effective management of software development projects.
Introduction
The role played by software in the common life have given both the academic and industry leaders increased thrust on its quality. The lessons learned from the very poor performance of much acclaimed softwares along with the very costly disasters this have caused have also prompted industry leaders to give priority in the quality of software development process. It is also reported that nearly 32 percent of the software projects are left incomplete and only 16 percent of the projects are completed within the specified time and allocated budget.
In the continued effort to find an effective solution for the quality of software projects developed by the industries , have resulted in a unanimous conclusion that this could be achieved only by evolving management systems that is capable to ensure the required quality.
Different systems proposed for the Software Process Improvement (SPI) are CMI model , CMMI model or ISO standards etc. The efforts made to operationalize these models or standards have resulted in the development of high quality software besides reducing the cost and time and improving the productivity. Most of the managers involved with the implementation of these systems demand for supporting them on how to implement these modules rather than identifying what module would be of actual help to them.
Every software development activity involves a complex network of relationship between the projects, people, management and the environment. And hence it require a detailed preparation of plan to complete the specific requirements with respect to the available resources for a particular task. The activities would start with the step called ‘requirement analysis” which is a very important part of software engineering.
The proper identification of the requirement by the software developers is necessary to design the most appropriate solution. Project monitoring and control is a very vital component in the project management process to avoid any significant deviation from the set objectives. The software development process any where on the world consists of numerous risk elements also. Software risk management refers to the process of measuring risk and incorporating appropriate means to mitigate its impact. This process is an integral component of all software development programmes.
The software development processes are addressed from the viewpoints of three important groups – project managers, software developers and customers. The problems encountered by the above mentioned groups varies significantly. All organization must need to have well structured operational systems capable of delivering the desired objectives by properly synchronizing the three major groups. CMMI model is the most recent among such initiatives towards developing effective software implementation process by effective coordination of all groups involved in the development.
CMMI (Capability Maturity Model Integrated) is frame work for operation of various processes in the software development cycle in order to make the operations very efficient. This frame work is proposed for ensuring quality practices in software development by giving appropriate guidance to the traditional practices and giving necessary points of assessments in current projects.
The important reasons usually stated for implementing the CMMI are
- Maintain the quality of process related to acquire, develop and maintain the software projects.
- Improvement in the product and service quality as a result of better processes,
- Aligning with business objectives.
Proper implementation of CMMI practices helps the organization to avail following benefits.
Appropriate linkages of the engineering and management process with the business goals.
To ensure that the products developed or services delivered meet the customer satisfaction.
Incorporating various invaluable lessons learned from the different areas of best practices in the industry like measurement division, supply chain management , logistics management etc.
Implementation more matured practices that are very robust.
Complying with various standardisation measures like ISO standards and identifying additional functional areas that need to be incorporated into this operation.
This report contains a detailed survey of CMMI framework and its significance in the software project management. The survey would help to understand the role played by CMMI maturity models in software development process and highlights the domains that necessitates serious attention.
Background for CMMI
The software process improvement (SPI) is a very expensive and long term programmme and it demands considerable input of resources – human effort, time and money. The major hurdle being observed for the effective implementation of the SPIs is the lack of importance given for the human factors involved. It is very important to consider the views, perception and opinions regarding the positive and negative influence of the strategies chosen in a particular method of implementation.
It is also observed that the advancement made in the development various models like CMMI has not been appropriately supported by the developing different strategies for their effective implementation. Thus the current problems are not in the availability of models or standards but a proper strategy for their implementation. Also, majority of software practitioners opinion that unless their role in various stages are made significant along with their training and mentoring would be very helpful in improving the success rate of the implementation.
Proper and effective reviews, senior management commitment and creating process action teams are also identified as very vital components required in the process of implementation of maturity models. Among the factors that act as barriers for the implementation activities are organizational politics, lack of experienced staff, inadequate awareness and also very less support extended to the team. Thus inorder to make the SPIs very successful it is inevitable to get the support of management and practitioners which demands extensive awareness creation programme within organizations.
In an effort to provide enough flexibility in the maturity models for successful implementation following three factors are identified as very important – maturity stage dimension, critical success factors and assessment dimension. Each maturity level in any SPI process like CMMI refers to the clearly designed step that hepls the organization to achieve a mature implementation process. Thus these levels would have the ability to define, manage and measure the process of implementation.
And thus, the CMMI models are divided into five different maturity levels of implementation. CMMI applies similar concepts of process management principles as Six-Sigma or Total Quality Management (TQM). CMMI targets the comprehensive process quality management and Six Sigma targets specifically on improving the product quality and hence finally improving the customer satisfaction. As per the definition of Six Sigma , this refers to having less than 3.4 defects with one million associations with a product or service.
Six Sigma systems could be effectively integrated with the level 4 and level 5 maturity levels of CMM1 processes for ensuring better products and services. Where as, TQM involves improving customer satisfaction by improving suppliers input and organizational processes. The review of published literature clearly show that the though majority of companies were able to attain the level 1 and 2 in CMMI with reasonable level of comfort, those capable to achieve level 3 and above was not very high. Only 20 % of the industries could achieve the level 3 of the process while those in the level 4 and level 5 is only just 10 percent.
At level 3 the organizations process definitions include development and maintenance of organizations standard processes along with its accumulated assets like extensive process documentation and description of the product or service lifecycle. Using different approaches like simulations , engineers and managers could explore possible ways to identify different ways to improve the system. The organization operational performance are quantitatively assessed in the level 4 of the maturity model based on different parameters like product or service quality, process cycle time and development time etc.
Simulations would be of much help in identifying the key performance indicators required for improving their processes. The CMMI level 5 expects the organizations to develop competency for continued growth through organizational innovation and deployment. Identifying the possible improvements that could be incorporated in the technology and process to meet the quality and process performance objectives of the organization or identifying the measurable improvements that sustains the growth of the industry by systematically developing and deploying the process and technology. Similar if tried at this stage could significantly reduce the experimentation cost , helps to validate the business cases for new technologies or reduce risks associated with new initiatives for improvement like staffing policies etc.
In addition simulations could also bring it other advantages to the organization. Possibilty to complete morre projects thus resulting in higher return in investment, very quick success can ehnace the CMMI implementation processes across the organization, prevents organization from choosing wrong initiatives are the few. Looking into the success attained by other quality assurance systems like Six Sigma through simulation, which have incorporated simulation as the important approach in its latest approaches CMMI too could use this tool to improve its acceptance in the organizations.
CMMI Process
Incorporation of systems capable to ensure the development of software products has been the yardstick to differentiate the industries ability to undertake projects of different dimensions. Though there are numerous choices available in the form of process quality models, Capability “Maturity Model Integration (CMMI) is reported as the best choice available for adopting the framework for evaluating, benchmarking, and improving software processes.”
Capability Maturity Model Integration (CMMI) is a frame work developed for effectively managing the software development process to deliver high quality products and services. This framework followed a path of evolution starting as a ad-hoc to evolve into more mature and capable process having five separate levels process areas. Thus Capability maturity model integration (CMMI) is approach developed for organisations with the objective of making their functional processes very effective and efficient. This improvement could be across the a project, a division or covering entire organization.
The broad outcome is to integrate different organizational functions , that have been traditionally considered , by setting process improvement goals and priorities and giving appropriate guidance and insisting on guideline towards achieving the quality expectations of the processes. The benefits that any organizations can expect on adopting quality practices are : acquiring the know-how about the methods to develop and maintain required quality systems, achieving the business objective with better product and service quality and creation of attitude to ensure process quality as a vital step to achieve business objective.
CMMI is not a tool to help organization to achieve its strategic business goals rather it a powerful support system to improve the organizations capability to achieve its business objectives. Based on the ability of organization to achieve the respective levels shown in the table, they are rated as the organization holding corresponding CMMI levels. In order to improve the levels of CMMI ratings the organizations must target the reduction of development times, eliminate the possible defects, forecast the process performance and suitable design products and services to meet the need of the customer. They can migrate from one level to the higher one by appropriate improvements in the operational process and skipping one level in the process is not recommended as it could turn counterproductive.
Continuous and Staged Representation
The process area in each maturity model is usually represented by two types of levels. Staged representation and continuous representation are the two different types of representations adopted in CMMI framework. In the case of software CMM, the predefined set of process parameters are required to define improvement path in the maturity model. This is the main characteristics of a staged representation. The maturity level corresponds to the well-defined path laid towards attaining improved organizational process.
While in the case of SCEM or IPD-CM the above steps involves choosing any selected process and then improving it relatively with other process. This type of approache is referred to “as staged representation. The important factors involved in the selection of these representations are business, culture and legacy. A comparison of relative advantages between both processes is given in Table 1.
Table 1: Comparison of representations:
Levels of CMMI
The brief description of different levels involved in CMMI are given in Table 2.
Table 2: Process Areas by Maturity Level (staged representation).
Project Management Process
Rapid advancement in computing technologies and influences of globalization have drastically changed the strategies involved in managing the software development projects. The recent trends in the managing software projects are in “distributed, pervasive and collaborative environments” which have inherent complexities that are not addressed in traditional systems. Thus, newer methods and measures to support software project management are necessary for effective management of software projects. With tremendous growth experienced in the software sector the projects have been undertaken at different locations across the globe.
The results of two projects which were operated on 24 hour continuous development time , with appropriate interchange of tasks between the teams which are positioned geographically different locations are presented. In the project to development of business intelligent solution this approach was successfully adopted while in the other which was related to a web application product greater challenges were met. Authors are of the opinion that these types of projects must have specifically designed project statutory and not a retrofitted type from the conventional models or traditional development methodologies. Their final views are expressed as
“although it does introduce new variables into the development process, we believe that using the follow-the-sun approach for software development can work. Given the project characteristics listed previously, successful communication between team members and an overall project structure designed to fit the complexities of these projects, this approach can be a successful way to decrease overall project costs and shorten schedules, provided the methodology is specifically designed to facilitate project success. We believe that as more experience is gained in this area, such advantages will be used more and more as a competitive differentiator in the software development marketplace.”
In another attempt to explore ways improve the software management process software agent framework, which forms part of SPMSA model is chosen. A prototype is developed for the risk management function area of this model, which can be implemented. This would help to prove that the model is not just theoretical piece but agent framework that could be put into operation directly. The result obtained from this research have established that it could enable the software developers to meet market expectations, and produce projects with in the stipulated time frame and allocated budget and to users satisfaction.
Another set of improvements are made possible by combining simulation with CMMI is also reported. The simulation approaches are specifically suitable in the situations where production or service facilities where experimentation need to be carried out is too costly, presence of environments with legal or life threatening ramifications to adverse changes or if proposed facilities have not been made ready. Also these approaches are said to be more capable due to following reasons
“Possibility of more standardized evaluations in line with CMMI using new object oriented simulation capabilities which reduces the time to build models. Improved completion rate of projects resulting on better return on investments. Generation of faster success stories helps to generate momentum for CMMI initiatives. Prevents wrong decisions very early using the testing results.”
Constrains to CMMI Implementation
The study highlights why organizations don’t adopt CMMI guidelines based on the analysis undertaken on the sales data collected by Australian company. Small size of the organization, costly services, less time to spent to integrate new practices, investment made already in alternate systems to ensure process quality are the reasons identified as the reluctance to adopt CMMI. Though the organizations categorically state that it would not be feasible to adopt CMMI systems, they are silent of the “possible benefits that could emerge from adopting a better set of practices. This study was undertaken by considering huge sample of organizations of different sizes and the varied nature of projects handled by them.”
The impact of the cultural change among the employees of software industries is another important parameter effecting software project management process. Based on the study undertaken over a period of two years on two business units – Integrated Systems (ISY) and Airborne Systems (ASY) the results highlight how variation in culture across organizations can “influence software process improvements. Based on the research analysis authors proposes a set of lessons to manage cultural variation in SPI initiatives.”
- Lesson 1: Consider cultural variations proactively during SPI.
- Lesson 2: Apply clinical inquiry analysis to uncover cultural variations.
- Lesson 3: Determine feasibility of adopted approach to SPI.
- Lesson 4: Accommodate cultural variation into SPI approach.
In spite of the advancement s made in the development and promotion of software process improvement standards and models, with the most recent being CMMI, has not been matched by appropriate increase in the adoption of these standards. This has resulted in the limited success of many of the software process improvement implementation projects. The existing issue is the absence of an effective strategy to successfully implement these standards or models and not the lack of any standards or models. The authors have attempted to propose some solution to SPI practitioners to design effective implementation strategies of quality systems. According to authors:
“We have pulled together individual components under one Software Process Improvement Implementation Framework (SPI-IF) using a bottom-up approach already familiar to many practitioners and researchers. The framework is based on the results drawn from SPI literature and an empirical study we have carried out. In the design of SPI-IF, the concept of critical success factors (CSFs) was used and extended. Thirty-four CSF interviews were conducted with Australian practitioners. In addition, 50 research articles (published experience reports and case studies) were also selected and analyzed in order to identify factors that play positive or negative roles in SPI implementation. The SPI-IF provides a very practical structure with which to assess and implement SPI implementation initiatives. In order to evaluate SPI-IF, a practical evaluation scheme was undertaken. The evaluation results show that SPI-IF has potential to assist SPI practitioners in the design of effective SPI implementation initiatives. Thus, we recommend organizations to use SPI-IF in order to effectively design SPI implementation initiatives.”
Monitoring, Controlling and Coordinating Projects
Studies have been undertaken to understand how the project managers control and coordinate software development tasks and how this is affected by changing environments. It has been observed that the changes in various environment of action resulted in the increased organizational distance between the project team members. The cultural, structural and administrative distance observed between the project manager and the other members of the project team is referred to as ‘organizational distance’. It is also observed that there is tremendous increase in both outsourced and distributed software development projects which in turn demands better approaches in project management practices.
Though the ways of monitoring, controlling and coordination of software projects are well researched further investigations are necessary into the factors influencing the organizational distance on the outcome of software projects. Further steps in research must probe into the extend of effect the organizational distance could cause on different mechanisms involved in the management of software projects.
If detailed information are available on the ways in which changed both the project environments and the selection of project management mechanisms, a better responses to these situations could be devised. Such level of information could help us to identify the tools that could be developed to assist project management of outsourced and distributed projects. Since there is very limited information available on the concept of organizational distance, a new model is developed that includes the dimensions of distance that may be found in most of the outsourced or globally distributed projects. Another model that is tried incorporates the changes in the organizational distance with preferred choices of project management mechanisms.
The research analysis was undertaken by collecting necessary quantitative and qualitative information from project managers engaged in software development using different techniques of data collection. The studies have clearly shown that the project managers do not rely on a single mechanism to monitor, control or coordinate a software development project but usually adopt multiple mechanisms.
The primary objective of project monitoring mechanisms was to first detect the presence of any project problems and later to address to those problems effectively. These systems stands totally different from the monitoring units that are designed to provide all the information about both the existence and expected causes of project problems. The origin of control rests with the project control mechanisms created in the project management systems.
The constrains in the form of budget and time are imposed by the organization and used by the project manager to direct the project to reach the desired outcome. Though the choice of project management techniques did not show any difference between the co-located projects and distributed projects, the project coordination mechanisms showed dependencies of different types of techniques between the software development processes.
The most common approach in the project management was using a project work breakdown structure usually expressed in the project schedule. This would be able to resolve sequential and pooled resource dependencies while, the interactive mechanisms like co-location, conversations and meetings would resolve the mutual dependencies. Based on the research results available it is established that distributed and globally outsourced software development projects may encounter many difficulties that are not experienced by a fully co-located project. But it must be understood that the attention to those difficulties seems to lie with the project management systems implementation and not by their selection.
Software Risk Management
Managing risk involved in the software development process is another important area that need to be addressed appropriately. The investigation have been undertaken to assess the utility of life cycle management theory to study the how risk shall be avoided in bidding for software projects. The risk involvement in a software project could be in various forms in its life cycle. As numerous types of risk items exist, a detailed analysis on the possible risk response measures for different risk categories is undertaken. Based on the detailed research analysis a basic methodology for dynamic risk avoidance seems appropriate for tackle the risks that appear at different stages.
The time taken to reach the market is the most critical factor influencing the success of any software projects. A quick survey into the development cycles of different types of software projects clearly shows that large numbers of these projects had time and cost over runs and majority of them also had quality and usability problems. A research to identify the root cause of these problems revealed that the organizational issues , like a feeling of lack of touch between software developers and management, were most predominant among them. Another major risk being perceived by many of the software companies especially in India is the high turnover rate of employees
The executives and software managers often considered the risks to be associated with error in the estimating and project schedule planning, inaccuracies in the status reports and the external pressures. It is widely considered in the software development sector that risk management would incur additional cost and effort. The opinion that most of the software development practitioners have on risk factor is also equally surprising. They consider that that the risk management measures in software development sector would be inhibiting the creativity of most of the software developers.
Though the risk management is least practiced discipline among the software development groups , research into a sample of failed software projects have revealed that effective intervention and initiation of corrective steps could have averted the observed failure of the activity. The recent statistics have shown that those successfully implementing the risk management measures are able to complete their projects successfully in comparison with the others who shows indifference towards any risk management initiatives. Also as the software projects turn large and complex the effort to incorporate risk control measures often turn cumbersome.
Even after undergoing extensive exercises many of the software projects fail to have user expectations, commonly delivered late and mostly exceed the proposed budget. These reasons clearly show the importance of the need for Software Project Management (SPM) in many organizations. SPM involves “the management of all aspects and issues that are involved in the development of a software project, namely identification of scope and objectives, project development approaches, software effort and cost estimation, activity planning, monitoring and control, risk management, resource allocation and control, as well as managing contracts, teams of people and quality”.
The unique characteristics of the software projects are invisibility, complexity, conformity and flexibility. These aspects often contribute not only to the difficulty in managing software projects but also to the failure of such projects. The key areas of importance during software project management process are classified as core and facilitating functions. The factors like scope, time, cost and quality management are the important core functions that directly influence the specific project objectives.
Human resource management, communication, risk and procurement management are the different type of functions that represent the means through which different objectives of the project are met. The project management tools and techniques are spread across the above-mentioned knowledge areas. They are often used to assist team members and project managers to undertake the core and facilitating functions.
The importance of effective communication, like repetitive face to face meetings, across the development team teams is reported to generate information and communication channels and resulted in the successful completion of the projects. The earlier studies have also stressed the importance of iterative development methodologies to development mutual trust between the teams and resulting in the mitigation of some of the risks associated with global software development projects.
Further, a detailed introspection into the methodologies adopted for managing the unsuccessful projects have shown that the social factors are those causing the negative results rather than the tools and techniques relating to the product content. The incidence of project failures can be significantly reduced by having a prior understanding of project risks. Based on a detailed study undertaken software project risk was conceptualized along six different dimensions. Also Wallace et al conducted a cluster analysis using a questionnaire survey involving 507 software project managers to identify various low, medium, and high-risk software projects.
A detailed examination of risk dimensions by the authors across the levels revealed that even low risk projects have a high level of complexity risk. For high-risk projects, the risks are often identified as associated with the requirements, planning and control. Also examined were the influence of project scope, sourcing practices, and strategic orientation on project risk dimensions. Results suggested that project scope affected all the dimensions of risk, while the sourcing practices and strategic orientation had a very limited impact.
Risk Management in CMMI
The risk management is very important aspect of software project management and effective management of the risks involved determines the success and failure of the software project. The three specific goals addressed in the software project risk management process area are “to prepare for the risk management“, “identification and analysis of risks” and ’risk mitigation.” One of the important features of CMMI is the ability to equip the operational systems to effectively manage risks involved in the project. It is available from the literature that CMMI has a very comprehensive concept of risk management and is in compliance with popular risk models like the risk management model of Boehm and SRE, TRM and CRM of SEI.
The key steps involved in any risk management process are defining the strategy involved in the risk management, risk identification and analysis and managing and mitigating the impacts of identified risks. These steps could be simultaneously undertaken along with the project planning’ and ‘project monitoring and control’ process areas. Risk management process involves a very comprehensive structure having a systematic plan and having abilities to anticipate and mitigate risks. Various methods normally used for risk mitigation are prototyping, simulation, alternative designs and evolutionary development.
In CMMI , the risk management process is essential component in the key process area of integrated along the software development process in the third maturity level. Thus CMMI based Software Project Risk Management (SPRM) framework would help the organizations to make the software projects more certain and regular. Based on the CMMI model the risk management process has eight separate components which starts with an effective risk response planning.
The planning is followed by the identification and analysis of risks. Planning the response actions and tracking the response follows the next. Controlling risks, their communication and learning completes the list. Such an approach would also help to integrate both qualitative and quantitative methods of risk management. In an effort to develop tools to help software managers take proper risk decisions on various problems economic optimization model is said to be of very help.
Here, the software managers proposes variety of risk reduction measures often identified in the projects based on the exhaustive list available in the repository and those collected from discussion from experts. A cost benefit analysis of various options are undertaken to select the most cost effective option that reduces the risk probability or impact. CMMI framework also helps to addresses the human dimensions to risk management, which besides the extensive training also includes systems to ensure effective communication and commitment for its implementation. Finally, it could be said that risk management in CMMI is a continuous process that is executed throughout the software development life cycle.
Summary and conclusion
Inspite of having very efficient frame work to improve the development processes still the success rates of development initiatives are not very high. Even the organizations that have attained higher levels of maturity, numerous problems with the usability of products have been observed. Also various cultural and social barriers also acts as an impediment for successful process development units. CMMI need to be tailored well for the small scale software enterprises as they often felt the cost of experimentation with new types of process systems may not match the expected returns from the business operations.
Further with considerable amount of work getting distributed across the globe a clarity in the process operations is inevitable. The importance of CMMI in making the process efficient inorder to easily achieve the business goals is observed.
The potential use of simulation in CMMI implementation process would be of much help in small scale operations. With the help of simulation based methods the skeptism in accepting these approaches could be easily eliminated for more widespread acceptability of maturity levels in processes. The risk management operations also appreciates the acceptability of CMMI framework based initiatives.
Hence, future approaches would be directed towards elimination of barriers in accepting the CMMI framework for all types and sizes of software development units. Along with increasing the acceptability of CMMMI frameworks , impact of such principals on the usability of the product also must be given attention along with ensuring the maturity level of the organization would certainly enhances their credibility.
References
Chrissis, Mary Beth, Konrad, Mike, Shrum, Sandy. (2003). Guidelines for process integration and product improvement. Addison-Wesley Publishing. Web.
Cmmi benefits. (2008). Software Engineering Institute. Web.
CMMI forsystems engineering software engineering. (2000): Software Engineering Institute. Web.
Lanubile., Damian., & Oppenheimer. (2003). Global Software Development: Technical, Organizational, and Social Challenges: ACM SIGSOFT Software Engineering, 28(6),1-4. Web.
McBride, Tom. (2005). The use of project management mechanisms in software development and their relationship to organizational distance. Sydney. Web.
MacGregor., Hsieh., & Kruchten. Cultural Patterns in Software Process Mishaps: Incidents in Global Projects, Proceedings of the Workshop on Haman and Social Factors of Software Engineering. Web.
Miller, Martin., Vidal, J Francisco Pulgar., & Ferriin, David M. Achieving higher levels of cmmi maturity using simulation,1473 -1478. Web.
Muller, Sune Dueholm., Kraemmergaard, Pernille., & Mathiassen Lars. (2008). Managing cultural variation in software process improvement: A comparison for methods for subculture assessment. Informatics Research Group. Working Paper 1-2008-01. Web.
Niazi, Mahmood., Wilson, David., & Zowghi, Didar. (2005). A framework for assisting the design of effective software process improvement implementation strategies: Journal of Systems and Software, 78(2), 204-222. Web.
Nienaber., et al. Software agent technology supporting risk management in spm. Web.
Paasivaara, Maria. Communication needs practices and supporting strtuctures in Global Inter organizational Software developments projects. Web.
Prikladnicki, Rafael., Audy, Jorge., & Evaristo, Roberto. Requirements management in global software development: Preliminary findings from a case study in a sw-cmm context. Web.
Software agent technology for risk management in spm. (2008). Meraka Institute. : African Institute for Information and Communication Technology. Web.
Staples, Mahmood., et al. (2007). An exploratory study of why organization don’t adopt CMMI ?. Journal of systems and software, 80(6), 883 – 895. Web.
Treinan., & Miller Frost, James J. (2006). Following the sun: case studies in global software development: IBM Systems Journal. Web.
Wallace, Linda., Keil, Mark., &Rai, Arun. (2004), Understanding software project risk: a cluster analysis. Information and Management, 42(1), 115 – 125. Web.
Welcome to the CMMI Website. (2008). Software Engineering Institute. Web.
Xie, Gang., Zhang, Jinlong., & Lai. (2006). International Journal of Project Management, Risk avoidance in bidding for software projects based on life cycle management theory, 24(6), 516-521. Web.