An effective framework to improve the managerial activities in global software development

1 Introduction

Global Software Development (GSD) is the process in which a software system is developed across two or more nationally and continentally separated locations. The Software Development Life Cycle (SDLC) phases are also geographically distributed, and the stakeholders involved are socio-culturally different [1,2]. The interesting GSD is remarkably increasing for its marketable benefits. Many industries have incorporated this technique into their software development practices. As the process of GSD takes place at different geographical locations, the components of the software are also distributed across different locations depending upon the size of the project. The distributed phenomenon reduces the development time and cost of the product. Customers get the deliverables on time, increasing their trust pool. Modularization also reduces the complexity of the software system which helps the customers in better understanding of the system. There are minimum chances of error propagation. GSD prevails over the sense of responsibility for development among various stakeholders. Large and highly skilled labor can be accessed, reducing labor costs. Creative and innovative ideas can be exchanged. However, the management area of GSD still lacks standards and strict planning for the successful completion of software projects.

It is inappropriate or lacks coordination in the development stages which is caused by communication problems, different cultures around the globe, work distribution tasks, uncertainty in organizational process understanding, requirement and design, planning of the project, and continuation. Due to a lack of visualization and less communication, the requirements gathered are incorrect and insufficient. This poor management problem poses many other serious challenges among which the main issue faced in GSD is product quality. It becomes very expensive to remove big errors when the distributed modules are integrated, or the product comes into operation. It is very difficult to trace back the phases of the SDLC to find the problem area at which the problem occurred and rework it. For instance, the product may lack any feature or not behave the way that the customer wants when tested.

This problem can be generated due to any missing, incorrect, or insufficient requirement. There is also a possibility that the developer may have perceived the requirement in some other sense due to socio-cultural differences, lack of visualization, and communication. Hence, this problem will require more time and cost to be solved. The development process will be decelerated as it will have to backpropagate, and the trust pool of the customer will also decrease. Therefore, to avoid such issues, the GSD process requires such techniques through which it can be properly managed. Different Strategies have been adopted so far to overcome these challenges but still, there is a dire need to address these issues comprehensively. Successful management will ultimately lead to the successful development of the software product.

In the GSD environment, the system development has made information and knowledge sharing an important element among team members that must be well organized. Many organizations have implemented GSD environment that faces many management issues. In this study, we conducted a literature survey from different but relevant research papers based on the following three formalized questions:

RQ1: What are the different challenges related to GSD?

RQ2: What are the possible solutions for acquiring better project management?

RQ3: How can we improve the managerial activities during SDLC in GSD?

2 Literature review

Most of the project’s failure depends on unsuitable project management practices at team, company, and country levels. The methods and processes considered by developed countries differ from those of developing countries. The difference between vendor and client companies’ perspectives is another issue. One focuses on skilled workers and saves costs; however, the latter wants a lucrative business and achieving projects [3]. The business model framework can facilitate the software companies to improve the project management and development processes [2]. Knowledge management deals with the communication and synchronization management of information regarding a project which affects the progress of the development of the project. There is very rare coverage on joined retrieval of knowledge, the ability to get new sources of knowledge, and how to handle tacit knowledge.

Noor and Rana [4] focused on using a common repository of information sharing which abridges the communication and coordination gaps. To improve the quality of products in GSD, Sanjay Misra et al. [5] highlighted the importance of quality assurance. Quality can be managed and enhanced by tracing the change occurring and controlling them through the management of risks and then properly reporting them. Proper planning for the development of projects near the customers will decrease the differences in culture and language. The active and thorough working relationship between the development team members during the project reduces the issues of communication and coordination. Strategic choices should be adopted to reduce the issues, for instance, outsourcing the parts of a project should only be considered for the non-culturally sensitive ones. Team members need to be skilled in other languages to reduce the communication gap. The exchange of team members across different sites for short intervals of time can improve the working environment.

Organizations are trying to adopt different strategies and methods for combining component-based software and a global-based environment. The challenges [2,4,5,6,7,8,9] that were identified during the management of globally distributed component-based software i.e., lack of understanding of ownership, decency of components on each other, insufficient process for the changes in the requirement, the difference between the cultures of team members working in an organization, and tracking of tasks and bugs are some of the challenges that we found in component-based GSD. The identified challenges can be used as guidance while developing strategies for component-based software related to GSD.

Component software development is a convenient and easy way of developing the system from the existing components, instead of using it from scratch. With the risks of GSD development, there is also some risk associated with component-based software development. Insufficient knowledge of components and interdependency between them may give rise to other project management challenges [7]. Jain and Suman [6] proposed a framework that combines the Project Management Book of Knowledge areas for great management of globally distributed software, such as feasibility management for risk, scope, resource, virtual, and integration management are also included. Some of the factors that can help in completion of a project successfully are the involvement of customers, and project management techniques that will increase reliability, and reduce the development cost and time. Managing globally distributed projects is very complicated. Lack of planning, irresponsibility, delay in timing, and inexperience results in overall project failure.

The Four C [7,8] problems are identified related to the GSD challenges such as Communication, Coordination, Collaboration, and Control problems. Successful virtual management is based on four elements that are cohesion between teams, configuration, allocation of tasks between team members, and sometimes conflicts arise between team members which need to be resolved. A knowledge management framework for GSD may be comprised of a system school, cartographic school, engineering school, organization school, spatial school, and strategic school. However, it is an inefficient approach to cost management that may result in low quality output.

A virtual team can be established to carry out the project management life cycle (PMLC) activities in globally distributed projects [8]. Communication in project management can be improved if there is a limited time constraint between team members of an organization and the socio-cultural and geographical differences among the team members are properly understood. Good interpersonal sensitivity skills will result in high-quality projects. Acceptance of others’ opinions and expressing what the members really want are important factors for achieving a high-quality product. The uncertainty of project completion is high in a global software environment. Due to the lack of methods and techniques, many risks associated with practitioners can result in project failures.

Arumugam et al. [10] proposed a framework, multi-agent simulation model (MASM), from which two agents called Global practitioner and Organizational agent can assess the risk associated with the global practitioners. MASM can be applied to an environment where the overall cost of developing the model is relatively high. Knowledge management issues can be generated due to a lack of communication, coordination, collaboration, and poor documentation. People usually do not have working experience in distributed environments [11]. The management of change is difficult in distributed environments. Data stores such as cloud computing can help to manage change [12]. Medium and large-scale organizations have closely related challenges of the requirement change management in GSD [13]. Less attention has been given to successfully executing the process improvement activities in the GSD environment.

Information sharing, building strong relationships, more staff involvement, a sense of responsibility, and skilled human resources can help to incorporate Software Process Improvement (SPI) techniques [14,15]. Because of integration complexities and incompatibilities, GSD integration issue arises, which in the end results in delay, loss of quality, chaos, and extra cost for the project [16]. The basic reason behind the issue of GSD integration is that the remote teams do not understand the working environments of distributed teams [17].

The members of teams at different locations and belonging to different cultures of the world follow different quality models, processes, and development standards. However, integration can be achieved by staying organized, keeping track of the processes, identifying the problems or risk a bit earlier by using good risk management techniques, and identifying failure factors like the class related to the single cause. Requirements are essential to successfully implement software projects [12,18 19 20 21 22]. Any change in the requirement during the development stage is unpreventable.

However, the change in requirement in the co-location is quite easy whereas in different locations, where the stakeholders are distributed across multiple locations, is difficult. This led to many challenges i.e., communication, collaboration, control, managing the central repository, and change in an efficient and effective manner [23,24]. Therefore, cloud computing can be applied to mitigate the challenges. The research [25,26] presents an improved framework called Reliability-centered maintenance (RCM) in GSD. The goal is to manage the requirement change in an efficient and effective manner. The proposed framework focuses on the processes of RCM and concerns of GSD are minimized.

During the literature analysis, the existing strategies to enhance project management activities in GSD were found to contain certain loopholes which need to be filled with further research in this domain [18]. Researchers and practitioners were invited to provide a more comprehensive solution to cover all the aspects of management.

3 Proposed framework

Many tools and techniques are accessible in the present world which addresses one or more than one problem but not all. As a solution to this, we have proposed a comprehensive framework that can help the teams to work more efficiently in a GSD environment. It can help to properly manage the managerial activities during the software development lifecycle and thus, reduce the failure rate which can be very expensive to afford if occurs at the end of SDLC. The proposed framework, as illustrated in Figure 1, has integrated multiple modules along with tools to provide a more comprehensive solution. There are six major modules, each dealing with a specific problem of managing activities in GSD. Each module contains important key elements which must be taken into account by the project management team. We have also associated the quality control module with project management modules to keep track of the quality of the product. For more accurate and efficient data management, all these modules are linked to a central data store.

Figure 1

The proposed framework to improve the managerial activities in GSD.

4 Evaluation of the proposed framework

To evaluate our proposed framework, we conducted an online survey from different IT-based companies across the world belonging to two categories i.e., GSD companies and local development (LD) companies. We received 48 responses. 62.5% of the companies belonged to GSD category involving near shore outsourcing and offshore outsourcing whereas 29.9% of companies were developing software locally involving onshore outsourcing. About 81.3% of the companies were multi-sited. The percentage of the companies that participated in GSD projects was 75% of all respondents. The outsourcing or multi-sourcing partner countries for the abovementioned categories are shown in Figure 2.

Figure 2

The outsourcing or multi-sourcing partner countries for GSD and LD companies.

Most of the companies were following Agile process model because Agile methodologies are more flexible than traditional ones in terms of change management. Agile helps in good management, rapid development, and accepts change in requirements at any stage of SDLC. Also, developers and testers can constantly interact with each other. A reliable software system can be developed in a shorter time. Few companies were following Scrum-based Agile methodology due to distributed chunks of functionalities among different teams. Scrum is the fast way of development when working on big projects and having to deliver on short time. Working in sprints, continuous meetings identify changes more easily which can be catered in future sprints. V-model was also adopted in few companies as testing across each phase of SDLC can identify errors early which saves time and cost.

The conceptual Model of GSD is shown in Figure 3.

Figure 3

Conceptual model of GSD.

As the development processes, developers, and information and technology are distributed, in order to identify certain challenges that are faced in such distributed project management, a checklist of possible challenges was provided in the survey form [33,34]. The results are shown in Figure 4.

Figure 4

Challenges faced by GSD companies GSD environment.

Most of the issues that were found during the survey are related to coordination, project planning, and communication and collaboration. Incorporating testing in each phase of SDLC is considered to be expensive and time consuming, but if it is incorporated, it has more benefits such that the end product will be more reliable. As there will be less failures in real-time, company will gain people’s trust. Fixing errors at the end is more expensive and time consuming than identifying and fixing them at an early stage with less efforts.

About 97.9% of the respondents, as illustrated in Figure 5, considered incorporating testing at each phase of SDLC as an effective approach. For instance, unit testing is mandatory to satisfy use cases. Therefore, we have proposed an idea of using hybrid model (combination of Agile and V&V). This hybrid approach can be effective for GSD. Having centralized data stores will enforce unified access knowledge and effective management.

Figure 5

Effectiveness of incorporating testing in each phase of SDLC.

About 91.7% of respondents, as illustrated in Figure 6, considered our idea of having centralized data stores as an effective approach. A few respondents were of the view that data centralization can be dangerous and vulnerable to threats such as security breaches. However, data stores can be protected by incorporating the best possible security measures.

Figure 6

Effectiveness of having centralized data stores.

The results of assessing the importance of other incorporated key elements in our proposed framework are shown in Figures 7–13.

Figure 7

Key elements that companies incorporate for quality control.

Figure 8

Management tools for effective management process.

Figure 9

Key elements that company incorporate for effective communication and collaboration.

Figure 10

Key elements that companies incorporate for effective coordination.

Figure 11

Key-elements that companies incorporate for effective change control management.

Figure 12

Key-elements that companies incorporate for effective cost estimation.

Figure 13

Key-elements that companies incorporate for effective effort estimate.

5 Conclusion and future work

GSD is widely employed in various organizations where specific expertise is essential to address specialized software development and support needs. The GSD industry represents a distinctive type of business that has exerted significant influence over the past several years, necessitating access to expert resources and prompt availability. The increasing reliance on software engineering has led to a transformation in contemporary software development practices.

In the realm of software development, project teams are composed of individuals from diverse geographical locations and cultural backgrounds. This setup offers numerous advantages, including continuous development operations, cost-effectiveness, and round-the-clock progress. The pivotal technologies in GSD are communication and collaboration tools, which play a central role in the functioning of distributed teams. However, these tools come with inherent challenges that need to be explored further. Notably, these technologies must effectively manage requirements, track project progress, and ensure that milestones are comprehended and agreed upon by all members of the GSD team. Eliminating irrelevant project activities that hinder project development progress is crucial.

This research proposes a framework designed to address management issues in GSD comprehensively. The objective is to offer a more holistic and robust solution to the challenges faced in this domain.

In the future, the focus of the researchers will be to bridge any gaps that exist in the current study. They aim to implement the proposed framework in various IT companies to attain more accurate results, utilizing data mining techniques. By doing so, they seek to enhance the effectiveness and efficiency of GSD in real-world scenarios, fostering better collaboration, streamlined processes, and successful project outcomes.