The impact of Artificial Intelligence & Machine Learning in
revolutionizing Software Project Management across Projects &
Products
Sameer Ali Khan
2167123
MSPM
IRS-2
Sir Afzal Ahmed
SZABIST Karachi Campus
1
Contents
1. Introduction 2
2. Literature Review: 3
3. Research Design & Methods: 6
4. References: 7
The impact of Artificial Intelligence & Machine Learning in revolutionizing Software
Project Management across Projects & Products
1. Introduction
How skillfully a software project is handled has a significant impact on its success. This
comprises essential tasks like determining the amount of work necessary to complete the
software project, developing a software project schedule that includes resource allocation,
managing project risks, keeping track of progress, etc. Thus, improper management of such
operations could have negative effects on software companies. Software project management,
however, is also a very difficult process that combines numerous business and human aspects, as
well as competing goals. When dealing with medium- to large-sized software projects, these
difficulties are increased. In these situations, artificial intelligence has the potential to be a great
help to software project managers, empowering them to make better management choices.
Project performance activities place a high value on project management planning and
assessment. It is difficult to manage a project effectively without a reasonable and logical plan.
This study will provide a thorough analysis of numerous studies on the use of machine learning
in software project management. A thorough literature review of (1) machine learning, (2)
software project management, and (3) methodologies from three major libraries—Web Science,
Science Directs, and IEEE Explore—is also presented in this study.
2
The focus would include studies on the phases and tests that are the parameters used in machine-
learning management and the final classes of the results from the study, contribution of studies in
the production, and promotion of machine-learning project prediction. The second category
consists of papers that are based on machine-learning methods and strategies used on projects. In
addition, our contribution provides a background and a wider viewpoint that are crucial for
possible project risk management efforts. We have demonstrated that project risk assessment
using machine learning is more effective at minimizing project losses, increasing the likelihood
that the project will succeed, offering a different way to effectively lower the probabilities of
project failure, and raising the output ratio.
Work on project monitoring and control is crucial and difficult. Any organization's most valuable
resource is its human capital, which guarantees the achievement of the company's goals and
objectives. To increase production, managers must organize and manage their teams. Employees
can contribute effectively and help an organization flourish if it has excellent human resources.
The talents and competency of an individual determine the role that process is assigned. The
manager will have an advantage when giving a new assignment if he is aware of these data about
the employee. According to the results of the previous study, managers spend a lot of time
allocating resources, which is done using their expertise.
Q1: What can be learned about software project management using machine learning approaches
from the academic material that is currently available?
Q2: What are the current research gaps for the area of software project management?
Q3: What is prediction metrics, and how accurate are they currently, as shown by various
estimating techniques?
Q4: Can we effectively lower the project failure rates by using feature transformation and feature
selection to develop more accurate machine learning-based models?
2. Literature Review:
Project managers have challenges in sustaining a software project's sustainability and increasing
its efficiency. Lack of knowledge, skills, resources, and technology during project
implementation are typically the causes of project failure [1,2,3].
3
By using either a mathematical methodology, such as linear regression and study of association,
or machine learning (ML) approaches, such as Artificial Network Network (ANN) and Support
Vector Machines (SVM), the knowledge obtained from historical project data sets can be used
for the development of predictive models. In order to foresee the project's future, predictive
approaches offer a method that is concentrated on both historical and current project information.
Since there are so many ML algorithms, not all of them have been thoroughly researched.
The purpose of using automated projects, problems with evaluating project management, and the
development of ML methodology are all addressed by the literature's conclusions. Evaluation of
the empirical findings is planned. Although there are extensive discussions about how to quantify
project improvements, project literature does highlight the project's success and failure. Project
success is judged differently depending on how it is seen [4]. Project success and project
performance factors are distinguished by Hughes and other members of the Project Management
Institute (PMI) [5] [6].
To determine if a project is successful or unsuccessful, the project progress thresholds are
assessed. The development of the project also takes feedback into account. A successful project
completed within a predetermined project length has historically been differentiated by the
delivery of the required results and the usage of the chosen resources [7]. Initiatives that
successfully meet the stakeholder's project goals, criteria, and ambitions are identified by PMI
[6]. The effects of the traditional objective outcome metrics, such as project expense (above and
below budget), project time (early, on or at late), and project outcomes output (with less or better
than required properties and functions), are discussed by researchers such as Como Aladwani
[8], Cates and Mollaghasemi [9], Parsons [10], and Rosenfeld [11].
Data Science and Artificial Intelligence have both been around since the 1950s, but only recently
have they seen a rise in popularity due to the increase of technology's capabilities. Further to this,
the main difference between Data Science and Artificial Intelligence is how they are applied; the
former is a process, whereas the latter is concerned with the use of datasets to guide decision-
making processes (Russell & Norvig, 2016). Both Deep Learning and Machine Learning are
4
subsets of artificial intelligence, however Deep Learning is more sophisticated than Machine
Learning (Goodfellow et al., 2016).
The roles of various sorts of data scientists are being promoted as the applications and uses of
data science processes spread. As time goes forward, humans when technology takes on easier
and more routine jobs, merging into higher-level roles. This requires a variety of functions that
are made required by the different steps and data types. Applying procedure Statisticians, spatial
data scientists, and business analysts are examples of the different roles. Among others,
practitioners, software programmers, and actuarial scientists. A large number of Statistics,
mathematics, and other subjects related to these positions are part of the equivalent prerequisite
along with others, examples include algebra, software engineering, and data intuition. The
elevated Global connectivity has also led to an increase in the amount of data available, which in
turn produces an environment that is enriching.
As current research broadens the boundaries and restrictions of these technologies, including the
capacity they can be utilized for, artificial intelligence, which manifests itself in the applications
of data science, is becoming increasingly common. To gather knowledge and information, one
example is to determine how Artificial Intelligence and Natural Language Processing can be
used on collaborative and semi-structured websites like Wikipedia (Hovy et al., 2013). It has
been estimated that there is a 50% chance that Artificial Intelligence may surpass human abilities
in as short as 45 years as this type of research becomes more prevalent (Grace et al., 2018).
In recent years, artificial intelligence has become more prevalent and pervasive across a variety
of industries (Wang, 2019). The project management sector in particular has seen a rise in
popularity, which may be attributed to the numerous advantages offered by its applications.
Support, accuracy, understanding, and strategy are a few of these advantages, along with the
elimination of information bias, the use of emotional intelligence, and creativity (Munir, 2019).
Artificial intelligence, a crucial application of data science, strongly interacts with statistics and
involves the capacity to use critical thinking in decision-making processes by posing the relevant
queries and using the proper delivery techniques (Rose, 2016).
5
Artificial intelligence applications are growing, evolving, and getting more sophisticated with
time. Project information, project tasks, the critical path approach, and program evaluation and
review techniques were the core topics of the early research projects on applications of artificial
intelligence (Foster, 1988; Levitt et al., 1988). But over time, its definition has expanded to
embrace automation.
3. Research Design & Methods:
This research will employ a comprehensive mixed-methods approach, encompassing both
quantitative and qualitative data collection strategies, to meticulously investigate the impact of
Artificial Intelligence (AI) and Machine Learning (ML) on software project management. Data
will be sourced from various channels, including:
Surveys: In order to gain deeper insights into the utilization of AI and ML in software project
management, an online survey will be administered to project managers and team members. The
survey instrument will be executed using Google Forms Sheets, which is regarded as a highly
efficient and user-friendly platform for data collection. This approach is chosen with the primary
aim of mitigating the potential risks associated with human error in data management.
The administration and compilation of survey responses will be orchestrated through Google
Forms, a robust tool for collecting feedback from a diverse pool of respondents. Leveraging
Google Forms enables the seamless management of data acquisition. The survey will be
disseminated through a panel provider, ensuring that respondents across various backgrounds
and contexts complete the form.
To ensure the robustness of the study, a rigorous sampling approach will be implemented. Both
an experimental group and a control group, representing distinct scenarios, will be sampled
equitably. Gender balance (50/50) will be meticulously maintained within the sample, thereby
fostering gender parity. Additionally, the age spectrum, ranging from 18 to 50 years, will be
judiciously considered in the sampling process. The subsequent section will elucidate the
statistical methodologies employed in the analysis of the gathered sample.
6
Profound, in-depth interviews will be conducted with key stakeholders actively involved in
software projects that have seamlessly integrated AI and ML technologies into their workflows.
These interviews will serve as a pivotal source of qualitative data, providing invaluable insights
into the experiences and perspectives of those at the forefront of AI and ML adoption in software
project management.
A meticulously selected array of software projects and products that have harnessed the power of
AI and ML will be subjected to in-depth case studies. This approach aims to holistically
understand and elucidate the outcomes, challenges, and lessons derived from their integration of
AI and ML technologies.
In summary, this study will adopt a multifaceted research design, combining the strengths of
quantitative surveys, qualitative interviews, and comprehensive case studies to provide a holistic
perspective on the transformative influence of AI and ML in software project management.
4. References:
1. Oun, T.A.; Blackburn, T.D.; Olson, B.A.; Blessner, P. An enterprise-wide knowledge
management approach to project management. Eng. Manag. J. 2016.
2. Maimone, C. Good Enough Project Management Practices for Researcher Support
Projects. In Proceedings of the Practice and Experience in Advanced Research
Computing on Rise of the Machines (Learning), Chicago, IL, USA, 28 July–1 August
2019.
3. Saleem, N. Empirical analysis of critical success factors for project management in global
software development. In Proceedings of the 2019 ACM/IEEE 14th International
Conference on Global Software Engineering (ICGSE), Montreal, QC, Canada, 25–26
May 2019.
4. Gemünden, H.G. Success factors of global new product development programs, the
definition of project success, knowledge sharing, and special issues of project
management journal. Proj. Manag. J. 2015.
5. Project Management Institute. Guide to the Project Management Body of Knowledge
(Pmbok Guide); Project Management Institute: Newtown Square, PA, USA, 2013.
6. Branscombe, M. (2018). How AI could revolutionize project management. CIO
Australia.
7
7. Munir, M. (2019). How Artificial Intelligence can help Project Managers? Global Journal
of Management And Business Research.
8. Nayab, N. (2019). An Overview Of Common Data Analysis Techniques Used In Project
Management. Bright Hub Project Management. Retrieved from
9. Nellutla, V. (2018). How Can Project Managers Use Data Science? Data Science Central.
10. Pedroso, M. (2017). Application of Machine Learning Techniques in Project
Management Tools.
11. Ram, J. (2017). What Role Does Big Data Have In Shaping The Future Of Project
Management.
12. Russell, S. J., & Norvig, P. (2016). Artificial intelligence: a modern approach: Malaysia;
Pearson Education Limited.
13. Sanchez, F., Micaelli, J.-P., Bonjour, E., & Monticolo, D. (2019). A Step for Improving
the Transition Between Traditional Project Management to Agile Project Management
Using a Project Management Maturity Model. The Journal of Modern Project
Management, 7(1).
14. Srinidhi, S. (2019). Data Science vs. Artificial Intelligence vs. Machine Learning vs.
Deep Learning.
15. Wang, Q. (2019). How to apply AI technology in Project Management.
16. Whyte, J., Stasis, A., & Lindkvist, C. (2016). Managing change in the delivery of
complex projects:
