CC BY
0
STUDY OF ARTIFICIAL INTELLIGENCE MODELS FOR BIG DATA ANALYSIS IN
PROJECT MANAGEMENT
D. Pshichenko, Associate Professor
National Research University Higher School of Economics (Russia, Moscow)
DOI:10.24412/2500-1000-2024-8-3-180-185
Abstract. This study explores the application of artificial intelligence (AI) and machine learning (ML) models for big data analysis in project management. By leveraging specific ML algorithms such as decision trees, random forests, support vector machines, neural networks, k-means clustering, gradient boosting, and natural language processing, project management practices are significantly enhanced. These technologies improve decision-making, resource allocation, and risk management. The implementation of these models involves addressing technical challenges, ensuring data quality, and adhering to ethical and privacy standards. This research provides an understanding of the transformative potential of AI and ML in optimizing project management.
Keywords: artificial intelligence (AI), machine learning (ML), big data, project management, decision trees, neural networks, risk management.
The integration of artificial intelligence (AI) and machine learning (ML) technologies is reshaping the landscape of project management. The rise of big data, characterized by its significant volume, velocity, variety, and veracity, presents both opportunities and challenges for project managers. Utilization of this data can lead to enhanced decision-making and more efficient project outcomes, but it requires advanced analytical tools and methods.
Big data offers a wealth of insights that, if harnessed correctly, can transform project management practices. However, the complexity and scale of these datasets often render traditional analytical approaches inapplicable. In such cases, AI and ML technologies are applied. It provides powerful algorithms and models capable of processing and analyzing large datasets, uncovering patterns, and generating predictive insights that are beyond the reach of conventional methods. The primary goal of this article is to explore how specific ML algorithms can be applied to big data analysis within the context of project management.
Theoretical foundations of AI and big data in project management
Technologies such as AI and ML are pivotal in driving advancements across various fields, including project management. AI en-
compasses a range of technologies designed to simulate human cognitive processes, enabling machines to perform tasks that typically require human intelligence. This includes capabilities like learning, reasoning, problemsolving, and understanding natural language. Machine learning, a subset of AI, involves the development of algorithms that allow computers to learn from and make predictions based on data. It focuses on creating models that can automatically improve through experience, thereby enhancing their performance over time.
In 2024, the AI market is estimated to be worth more than 184 billion USD, a significant increase of nearly 50 billion compared to 2023. This robust growth is expected to continue, with the market projected to exceed 826 billion USD by 2030 [1].
Big data, characterized by its volume, velocity, variety, and veracity, plays a crucial role in project management. The volume refers to the massive amount of data generated daily, while velocity indicates the speed at which new data is created and processed. According to forecasts, the total volume of data created, collected, copied, and consumed worldwide is expected to reach 180 zettabytes by 2025 [2]. Variety encompasses the different types of data, ranging from structured to
unstructured formats, and veracity pertains to the quality and reliability of the data.
The integration of AI and ML into project management allows for more accurate risk assessments, better resource allocation, and more efficient scheduling. These technologies enable the analysis of historical data to predict future project risks and outcomes, thus facilitating proactive decision-making. For instance, ML algorithms can analyze past project data to identify patterns and trends that may indicate potential issues in current or future projects. This predictive capability is invaluable for project managers, allowing them to mitigate risks before they materialize.
Big Data analytics in project management
The vast amount of data generated during project lifecycles, including project plans, budgets, schedules, and performance metrics, can be harnessed to gain insights that drive efficiency and effectiveness. By leveraging
Table 1. Key techniques for data collection, ment [4, 5]
Big data analytics is a powerful tool for enhancing project management practices. By effectively collecting, processing, and analyzing vast amounts of data, project managers can make more informed decisions, optimize resource utilization, and improve overall project performance. The integration of advanced analytical techniques into project management processes is essential for navigating the
big data, project managers can anticipate potential issues, mitigate risks, and make data-driven decisions that align with strategic objectives.
Data relevant to project management comes from various sources and in different formats [3]. Key data sources include internal systems such as project management software, enterprise resource planning (ERP) systems, and customer relationship management (CRM) platforms. External sources, such as market trends, industry benchmarks, and stakeholder feedback, also provide valuable information. The types of data can be broadly categorized into structured data, such as numerical and categorical data stored in databases, and unstructured data, such as emails, documents, and social media posts. Table 1 provides an overview of key techniques for data collection, processing, and analysis in project management.
processing, and analysis in project manage-
complexities of modern project environments and achieving successful project outcomes.
AI and ML algorithms in project management
The application of AI and ML algorithms in project management is transforming how projects are planned, executed, and monitored. These advanced technologies enable project managers to handle complex data, predict potential risks, optimize resource allo-
Technique Description Application in project management
Automated data capture Use of software tools to automatically collect data from project management systems Real-time tracking of project metrics, reducing manual entry errors
Manual data entry Human input of data into systems Capturing qualitative insights and unstructured data
API integration Connecting to external data sources through API Incorporating market trends and external benchmarks into project analysis
Data cleaning Removing errors and inconsistencies from data Ensuring accuracy and reliability of project performance reports
Data integration Combining data from multiple sources Creating comprehensive datasets for holistic project analysis
Data transformation Converting data into suitable formats foi analysis Preparing data for advanced analytics and visualization
Descriptive analytics Summarizing past project data Understanding historical performance and identifying trends
Predictive analytics Forecasting future outcomes using models and algorithms Anticipating risks and planning mitigation strategies
Prescriptive analytics Recommending actions based on predictive insights Optimizing project plans and resource allocation for better outcomes
cation, and improve overall project efficiency [6].
Decision trees are an important ML algorithm utilized in project management for their simplicity and interpretability. These algorithms enable managers to make informed decisions by mapping out possible outcomes and their associated risks. Decision trees are particularly effective in risk assessment and management, allowing for the identification
and mitigation of potential issues before they escalate [7]. By evaluating historical project data, decision trees can predict the likelihood of project delays, budget overruns, and other critical factors, thereby enabling proactive measures to be taken.
Random forest, an ensemble learning method that builds multiple decision trees and merges their results, enhances predictive accuracy and robustness (fig. 1).
Dataset
.*.*.*..*.*.*. .v.*.
Decision Tree-1 Decision Tree-2 Decision Tree-N Result-1 Result -2 Result - N
I _i_J
-H Majority Voting / Averaging p-
T
Final Result
Fig. 1. Decision trees scheme
In project management, random forests are employed to optimize resource allocation, schedule planning, and cost estimation This algorithm's ability to handle large datasets and reduce overfitting makes it a powerful tool for analyzing complex project variables and dependencies. By integrating random forests into project management systems, organizations can achieve more reliable and accu-
rate forecasts, leading to better resource utilization and project efficiency.
Support vector machines (SVM) are another ML algorithm extensively applied in project management. SVM are particularly effective in classification tasks, making them ideal for categorizing project risks, prioritizing tasks, and ensuring quality control (fig. 2).
By analyzing project data, SVM can distinguish between high-risk and low-risk activities, enabling managers to allocate resources and attention accordingly. The high-
dimensional space capabilities of SVM make them suitable for projects with numerous variables and intricate relationships [8].
Neural networks, especially deep learning models, have transformed project management by offering advanced predictive analytics and pattern recognition capabilities. These algorithms excel in modeling complex, nonlinear relationships within large datasets. In project management, neural networks are used to forecast project performance, detect anomalies, and predict future trends. Their ability to learn from vast amounts of data al-
lows for more accurate and nuanced predictions, which are critical for long-term project planning and strategy development.
K-means clustering is a valuable unsupervised learning algorithm used to group similar data points into clusters. In project management, k-means clustering is employed to segment project tasks, identify project phases, and understand team dynamics (fig. 3).
Unlabeled Data
Labeled Clusters
-
• • • • • • • • ••* K-Means ( • ) ( Ve / \
• •• • • • X = Ce n froid- •J
Fig. 3. K-means clustering diagram
This technique helps in organizing projects into manageable segments, improving clarity and focus. By clustering similar tasks, managers can streamline workflows, allocate resources more effectively, and enhance team collaboration.
Gradient boosting, another ensemble learning technique, combines weak predictive models to create a strong overall model. This algorithm is highly effective in forecasting project timelines, enhancing resource planning, and improving project accuracy (fig. 4).
Fig. 4. Gradient boosting clustering diagram
Gradient boosting's iterative approach to model building allows for continuous improvement in prediction accuracy, making it a valuable tool for dynamic project environments.
Natural language processing (NLP) is
increasingly important in project management, particularly for analyzing unstructured text data [9]. NLP algorithms can process and interpret project documentation, stakeholder communications, and other text-heavy sources. This capability enables the extraction
of relevant insights, sentiment analysis, and the identification of potential issues or concerns. By leveraging NLP, project managers can enhance their understanding of project contexts and stakeholder expectations, leading to more informed and responsive management practices.
From the author's perspective, the integration of these algorithms into project management processes can significantly enhance decision-making, optimize resource utilization, and improve project outcomes. By leveraging
the strengths of each algorithm, project managers can address the multifaceted challenges of modern project environments more effectively.
Challenges and limitations of implementing AI and ML in project management
Implementing AI and ML algorithms in project management comes with a set of significant challenges and limitations that need to be addressed to harness their full potential. Technical challenges are paramount among these, as the deployment of AI and ML systems often requires advanced infrastructure and specialized expertise. The complexity of these algorithms necessitates robust computational power and extensive data storage capabilities [10]. Developing and maintaining AI models demand a high level of technical proficiency, which may not be readily available within all organizations. The integration of AI systems with existing project management tools also poses a challenge, requiring seamless interoperability and compatibility.
Data quality and integration issues further complicate the implementation of AI and ML in project management. The effectiveness of these technologies is heavily dependent on the quality of the data they process. Poor data quality, characterized by inaccuracies, incompleteness, and inconsistencies, can significantly undermine the performance of AI algorithms [11]. Integrating data from disparate sources, each with its own format and structure, is a complex task that can lead to data silos and fragmented insights. Ensuring that data is clean, well-structured, and harmonized is crucial for accurate analysis and reliable predictions. Effective data integration strategies are essential to provide a comprehensive and cohesive view of project data.
Ethical considerations and data privacy
protection are crucial issues that cannot be overlooked in the context of AI and ML implementation. AI models trained on biased
data can shape or amplify existing biases, leading to incorrect outcomes. Another important aspect is data privacy, as AI and ML applications often involve processing sensitive and personal information. Compliance with data protection regulations such as GDPR and CCPA is mandatory to safeguard individuals' privacy rights [12]. Implementing stringent data security measures and developing clear data governance policies are essential to protect against data breaches and unauthorized access.
While the integration of AI and ML in project management holds substantial promise, addressing the technical challenges, data quality and integration issues, and ethical considerations and data privacy concerns is critical for successful implementation [13]. Overcoming these hurdles requires a strategic approach that includes investing in the necessary infrastructure and expertise, ensuring high data quality and effective integration, and adhering to ethical standards and privacy regulations. By tackling these challenges head-on, organizations can fully leverage the transformative potential of AI and ML technologies to enhance project management practices and achieve superior outcomes.
Conclusion
Integrating AI and ML into project management significantly enhances decision-making, resource utilization, and overall project outcomes. Specific ML algorithms like decision trees, random forests, support vector machines, neural networks, k-means clustering, gradient boosting, and natural language processing improve various project management aspects. Addressing technical challenges, ensuring high data quality, and adhering to ethical and privacy standards are crucial for successful implementation. This strategic approach allows organizations to leverage AI and ML's full potential, leading to more efficient and effective project management practices.
References
1. Artificial intelligence (AI) market size worldwide from 2020 to 2030 / Statista. - URL: https://www.statista.com/forecasts/1474143/global-ai-market-size (date of application: 15.07.2024).
2. Nguyen D.K., Sermpinis G., Stasinakis C. Big data, artificial intelligence and machine learning: A transformative symbiosis in favour of financial technology // European Financial Management. - 2023. - Vol. 29, № 2. - P. 517-548.
3. Volume of data/information created, captured, copied, and consumed worldwide from 2010 to 2020, with forecasts from 2021 to 2025 / Statista. - URL: https://www.statista.com/statistics/871513/worldwide-data-created/ (date of application: 20.07.2024).
4. Bobunov A.Yu. Optimization of performance and reliability of financial applications through innovative testing methods // Innovacionnaya nauka. - 2024. - № 6-1. - P. 74-79.
5. Dubey R., Bryde D. J., Graham G., Foropon C., Kumari S., Gupta O. The role of alliance management, big data analytics and information visibility on new-product development capability // Annals of Operations Research. - 2024. - P. 1-25.
6. Ogarkov A. Application of big data analytics to improve business customer service // Inno-vacionnaya nauka. - 2024. - № 7-1. - P. 61-65.
7. Costa V. G., Pedreira C. E. Recent advances in decision trees: An updated survey // Artificial Intelligence Review. - 2023. - Vol. 56, № 5. - P. 4765-4800.
8. Ikotun A.M., Ezugwu A.E., Abualigah L., Abuhaija B., Heming J. K-means clustering algorithms: A comprehensive review, variants analysis, and advances in the era of big data // Information Sciences. - 2023. - Vol. 622. - P. 178-210.
9. Korostin O. Application of NLP technologies for data extraction from text messages in maritime logistics // The scientific heritage. - 2024. - № 141. - P 42-45.
10. Odejide O.A., Edunjobi T.E. AI in project management: exploring theoretical models for decision-making and risk management // Engineering Science & Technology Journal. - 2024. -Vol. 5, № 3. - P. 1072-1085.
11. Vial G., Cameron A. F., Giannelia T., Jiang J. Managing artificial intelligence projects: Key insights from an AI consulting firm // Information Systems Journal. - 2023. - Vol. 33, № 3. - P. 669-691.
12. Mozharovskii E. Evaluating retrieval-augmented generation (RAG) techniques in enhancing LMS for coding tasks // Universum: tekhnicheskie nauki: lectron. nauchn. zhurn. - 2024. -№ 6 (123). - URL: https://7universum.com/ru/tech/archive/item/17729.
13. Bukhtueva I. The role of AI in financial risk management and fraud detection // ISJ Theoretical & Applied Science. - 2024. - № 07(135). - P. 65-69.
ИССЛЕДОВАНИЕ МОДЕЛЕЙ ИСКУССТВЕННОГО ИНТЕЛЛЕКТА ДЛЯ АНАЛИЗА БОЛЬШИХ ДАННЫХ В УПРАВЛЕНИИ ПРОЕКТАМИ
Д. Пшиченко, доцент
Высшая школа бизнеса НИУ ВШЭ
(Россия, г. Москва)
Аннотация. В данном исследовании рассматривается применение моделей искусственного интеллекта (ИИ) и машинного обучения (МО) для анализа больших данных в управлении проектами. Использование специфических алгоритмов МО, таких как деревья решений, случайные леса, машины опорных векторов, нейронные сети, кластеризация методом ^средних, градиентный бустинг и обработка естественного языка, значительно улучшает практики управления проектами. Эти технологии способствуют улучшению процессов принятия решений, распределения ресурсов и управления рисками. Реализация данных моделей требует решения технических задач, обеспечения качества данных и соблюдения этических и конфиденциальных стандартов. Данное исследование предоставляет понимание трансформационного потенциала ИИ и МО в оптимизации управления проектами.
Ключевые слова: искусственный интеллект (ИИ), машинное обучение (МО), большие данные, управление проектами, деревья решений, нейронные сети, управление рисками.
