Comprehensive development of a model and a project management tool in landscape design based on artificial intelligence

Full Text

Introduction

The rapid development of digital technologies, especially in the field of artificial intelligence, provides new opportunities to improve the effectiveness of project management, which allows for more accurate planning, adaptation to changes and risk reduction. However, despite the increasing interest in the digitalization of processes in project activities, the issues of introducing artificial intelligence into the management of enterprises operating in the field of landscape design remain insufficiently studied.
The relevance of this research is related to the need to create new approaches to project management based on the use of intelligent algorithms that are able to process large amounts of data, identify hidden patterns and offer optimal solutions. Traditional management models often do not have the necessary flexibility and do not take into account the specifics of modern challenges faced by enterprises in this area. In this regard, the development of an adaptive model and a formalized management tool is a top priority for improving the competitiveness and sustainability of project activities.
The purpose of this article is to comprehensively develop a model and a project management tool in the field of landscape design using artificial intelligence methods. The research examines existing approaches to project management, analyzes the applied performance criteria, and formulates the author's decision-making algorithm based on artificial intelligence technologies. The proposed model is aimed at improving the accuracy of estimating design parameters, automating some management functions and reducing the influence of subjective factors in the decision-making process.

Literature review

Modern project management science offers a variety of solutions, from time-tested classical techniques to adaptive and digital strategies.
The research covers a wide range of topics, including the transformation of lean manufacturing in project management (Puntikov and Shikov [18]), comparative analysis of traditional and flexible methodologies (Schildt and Gareeva [21]), innovative project management taking into account foresight and sustainable development (Mironova et al. [17]), the introduction of project management in the public sector with an emphasis on risk assessment (Greenman [11] and Kuzmina [16]), project management problems in IT companies and ways to solve them (Borchin [1], Vakorin [2] and Sulkowski), an analysis of world experience in public administration (Kozlovskaya [14] and Rizoeva) and the relationship between change management and projects (Zhunisov [13]).
The theoretical foundation remains the work of Gantt [3], Taylor [19] and Fayol [20]. Practical guides such as DeMarco's Deadline [12] and Cohn's Agile [15] make complex concepts more accessible.
Standardization and regulatory framework: GOST standards provide a methodological framework for standardization of quality management and projects, especially in the context of digital transformation and artificial intelligence assessment [10].
The analyzed literature highlights the need to combine classical principles with digital innovations to create hybrid project management models applicable in various industries and at different levels.
In general, the sources reviewed provide a comprehensive understanding of current problems and solutions in the field of project management. Regulatory documents such as GOST R 54869-2011 [4], ISO 21500:2014 [6] and ISO 9001:2015 [7] establish requirements for the quality and structure of project processes, especially relevant in the era of digitalization and the introduction of artificial intelligence [8].

Materials and methods
 
In the literature under study, the authors use a variety of methodological approaches to analyze project management.
Analytical methods:
• Comparison of theory and practice: Many researchers (for example, Schildt, Gareeva, Greenman) They analyze project management by comparing theoretical concepts with real-world cases.
• Terminological analysis: Puntikov and Shikov focus on clarifying terminology, distinguishing between different approaches of lean management.
• Modeling and forecasting: Mironova and colleagues apply modeling of the project environment, using foresight methods and principles of industrial symbiosis to predict the development of projects.
• Functional and structural analysis: The classic works of Gantt, Taylor, and Fayol laid the foundation for functional and structural analysis, which is still used in process design.
• Comparative analysis: Zhunisov analyzes the interaction between change management and project management, and compares international and domestic management practices.
• Case study: Kuzmina uses the case study method to demonstrate the application of risk assessment in educational projects.
• Problem-oriented approach: Vakorin, Sulkowski, and Borchin identify weaknesses in IT project management using a problem-oriented approach.
• Instrumental methods: Cohn and DeMarco analyze Agile and SCRUM instrumental methods, supporting their conclusions with practical examples.
Normative and standardized methods:
 
• Regulatory and technical methods: GOST standards standardize quality assessment and management processes based on regulatory and technical methods [5].
• Artificial intelligence analytics and digital monitoring: Some standards, such as GOST R ISO/IEC 24029-2-2024 [9], involve the use of AI analytics and digital monitoring methods.
Other methods:
• Expert assessment: The expert assessment method is used, especially when introducing artificial intelligence into project management.
• A systematic approach: The publications follow a systematic approach that considers projects as a set of resources, processes, and goals.
• Inductive-deductive logic: The authors use inductive-deductive logic to deduce general patterns from particular cases.
In general, the research is aimed at adapting classical management models to modern conditions and integrating digital technologies, including artificial intelligence, into project activities.

The results of the study
 
Evaluating the effectiveness of a project is an important aspect for analyzing how much the chosen management model contributes to achieving the goals set. In traditional methodologies such as PMBOK (Project Management Body of Knowledge), several criteria are identified by which a decision is made on the success of a project.:
1. Meeting deadlines is an assessment of how well the actual completion of tasks meets the planned schedule. All deviations are recorded and, if necessary, a new plan is developed.
2. Compliance with the allocated budget — the discrepancies with the budget, the efficiency of resource use and the degree of overspending are analyzed.
3. The quality of the result is determined through the fulfillment of technical specifications, the absence of defects and compliance with customer expectations.
4. Customer satisfaction is a subjective but important indicator that encompasses emotional response, the degree of compliance with expectations, and the transparency of the team's actions.
5. Effective communication is critically important for projects involving multiple teams or disciplines. The level of interaction, coordination of solutions, and speed of response to changes are assessed.
6. Risk management is determined by how successfully potential threats are identified, their impact is reduced, and corrective actions are taken.
In practice, qualitative assessment methods such as document analysis, expert opinions, and questionnaires are used, as well as quantitative parameters, including the percentage of tasks completed on time, the cost performance index (CPI), and the schedule performance index (SPI).
Current trends also include integral indexes that combine multiple metrics into a single system, as well as metrics based on digital footprints that examine activity in online spaces (for example, the speed of making changes to a project, interaction with artificial intelligence tools, and the history of errors and edits).
It should be noted that the criteria for evaluating the effectiveness of project management are fixed in various regulatory documents. For example, GOST R 54869-2011 "Project management. Project Management Requirements" and GOST R ISO 21500-2014 "Project Management Guidelines" formalize the main parameters of project success, including deadlines, budget, quality and work with stakeholders.
However, even with a well-established set of criteria, there are serious limitations in their application, especially in the field of landscape design, which includes a high degree of complexity, multi-stage approach and a significant level of creativity.
Existing project evaluation methods have a number of significant drawbacks, especially in dynamic and innovative areas such as landscape design using artificial intelligence.
First, they often do not take into account the real variability and unpredictability of the project environment. Factors such as supply disruptions, weather anomalies, or changes in customer requirements have a significant impact on timing and quality, but are not always adequately reflected in the assessment.
Secondly, traditional criteria do not take into account the long-term consequences of mistakes made in the initial stages. Planning problems, inaccurate data on soil and water, and unsuitable plants may lead to the collapse of the project in the future, but the current assessment system does not anticipate this.
Thirdly, the existing models do not integrate the role of modern intelligent technologies. For example, if an artificial intelligence system warns about the risk of plant death, and the manager ignores this warning, the consequences of this decision are not reflected in the effectiveness assessment.
As a result, the traditional criteria:
• They do not take into account the cumulative effect of errors and their long-term consequences.
• Do not reflect dynamic risks and their impact on the project.
• They do not sufficiently evaluate the qualitative aspects of the decisions they make.
• Not adapted to the use of digital and intelligent management tools.
These limitations emphasize the need to review approaches to project evaluation and, more broadly, to restructure the entire management system. The new model should take into account not only quantitative indicators, but also qualitative parameters reflecting the complexity and versatility of the project environment.
Thus, the obvious limitations of traditional assessment methods, recognized both in scientific research and in regulatory documents, require the development of a fundamentally new approach capable of adequately evaluating modern projects, especially in creative, dynamic and high-tech fields.
Given the existing shortcomings in project management, this paper offers a model specifically designed for the digital environment. In addition, a performance assessment tool is presented that goes beyond standard indicators (time, budget, risks) and includes important aspects such as the likelihood of errors and the quality of the final product. The tool is based on a formula that integrates key project indicators into a single evaluation metric, which provides a more objective analysis and, as a result, more informed management decisions.
The use of artificial intelligence in landscape design project management can significantly improve overall efficiency by providing more accurate forecasts, optimal resource allocation, and automation of repetitive operations. The use of artificial intelligence at various stages, starting with the analysis of initial data and the assessment of the territory and ending with the generation of design solutions and 3D modeling, significantly reduces time costs and reduces the risk of errors.
The developed model (Fig. 1) demonstrates how artificial intelligence can be used to increase the sustainability, flexibility and adaptability of design solutions in the context of landscape architecture, making them more reliable and efficient.

Fig. 1. Project management model using artificial intelligence

Thus, the proposed model of project management in the field of landscape design using artificial intelligence is an effective tool for improving the quality of design solutions and improving management processes. The integration of artificial intelligence at key stages of the project lifecycle not only speeds up the completion of tasks and increases the accuracy of decisions, but also reduces the burden on specialists, reduces the likelihood of errors and increases resilience to risks and uncertainties related to the human aspect.
Given the multifaceted impact of artificial intelligence on project activities, a tool has also been created in this paper to evaluate its implementation. This tool is a formula that allows you to take into account key parameters such as the quality of the result, time costs, the level of project risks, error rate and the number of employees involved. The use of this formula makes it possible to objectively assess the feasibility of using artificial intelligence in specific conditions and is the basis for making informed management decisions.
To analyze the effectiveness of the implementation of artificial intelligence in project management, an integral indicator was developed.:
The presented formula takes into account changes in various aspects of work, such as task completion time (V), number of employees (R), probability of errors (E), costs (C) and quality of the final product (Q) both before and after the introduction of artificial intelligence technologies. The weighting coefficients K₁–k₅ allow you to adapt the formula to the specifics of each individual project.
As a result of testing of this proposed tool, efficiency data were obtained using the example of the considered suburban landscape design project (see Table 1).

Table 1 - Project data before and after the introduction of artificial intelligence

Parameters	Before the introduction of artificial intelligence	After the introduction of artificial intelligence	Changes
Project time	85 days	65 days	-20%
Number of employees	15 people	10 people	-33%
Number of errors	48	15	-68%
Project costs	10,750,000 rubles	8,050,000 rubles	-25%
Project quality	70%	85%	+15%

The introduction of artificial intelligence into project management has made it possible to achieve a significant increase in return on investment. A 20% reduction in project implementation time and a 33% optimization of staff resulted in a 25% reduction in costs. At the same time, the quality of projects improved by 15%, and the number of errors decreased by 68%, which confirms the effectiveness of artificial intelligence as a tool to increase productivity and reduce risks.
The effectiveness of the proposed solutions using artificial intelligence is estimated at 35% (Ieff ≈ 35%). This indicates a significant improvement in the project's performance after the introduction of artificial intelligence.
The created system provides a comprehensive assessment of the effectiveness of the project stages in five key parameters: time, cost, number of staff, probability of errors and quality of the final result. This integrated approach contributes to a more objective analysis of the project's implementation, identifying its strengths and identifying areas for improvement.
The formula for evaluating the effectiveness of implementing artificial intelligence in landscape design project management meets various regulatory requirements developed both in Russia and internationally. For example, GOST R ISO/IEC 25010-2015 describes the quality characteristics of software and digital systems, including parameters such as reliability, efficiency, and ease of maintenance. These characteristics are closely related to the error probability (e₀/e₁) and the quality of design solutions (q₀/q₁), which are reflected in the formula.
The proposed formula for evaluating efficiency, using variables V (time), R (resources) and C (costs), is based on the principles of quality management and projects, enshrined in the standards GOST R ISO 9001-2015 and GOST R ISO 10006-2005. Moreover, it takes into account modern requirements for the analysis and monitoring of the effectiveness of AI solutions, established in GOST R ISO/IEC 24029-2-2024 and GOST R ISO/IEC 20547-3-2024. The introduction of weighting coefficients (K₁–k₅) allows you to adapt the model to the specifics of each project, prioritizing and focusing on key parameters, which makes it meet recognized standards in the field of quality, digital design and project management.

Discussion and conclusion

With the rapid progress of digital technologies, project management in the field of landscape design requires adaptation to new challenges. An analysis of current approaches shows that traditional management methods are often insufficient to take into account all the key factors necessary for the successful implementation of projects, such as quality, risks, time costs, human factors and financial aspects.
Based on the study of scientific sources and current standards, it was found that modern criteria for evaluating the effectiveness of projects need to be supplemented and systematized. Existing regulatory documents, including GOST R ISO/IEC 25010-2015, GOST R ISO 9001-2015, GOST R ISO 10006-2005 and others, create a reliable methodological base, but do not fully take into account the specifics of the use of artificial intelligence in project management.
The proposed management model using artificial intelligence shows the potential for improving the adaptability, sustainability, and accuracy of design decisions. It aims to reduce the likelihood of errors, optimize resource usage, reduce lead times, and improve the overall quality of the final product. The developed formula for evaluating the effectiveness of the introduction of artificial intelligence allows for an objective and quantitative assessment of the results of the digital transformation of management, taking into account five main indicators: time, costs, number of employees, errors and quality.
The author's proposals were tested on the data of the suburban area project, which allowed us to make a comparison, which shows that the use of artificial intelligence in the project increases its effectiveness by about 35%.
The connection of each element of the formula with the corresponding GOST standards emphasizes its compliance with international and domestic standards, and also enhances the validity and practical applicability of the proposed approach.
Thus, the results of the conducted research confirm the expediency and relevance of using artificial intelligence in managing landscape design projects. The implementation of this model and assessment tool can become the basis for creating more flexible, intelligent and efficient design systems in the architecture and construction practice of the future.

About the authors

Roman Aleksandrovich Gordovskiy

National Research University "Moscow Power Engineering Institute"

Author for correspondence.
Email: romangordovskiy@mail.ru

Student of the Department of Management in Energy and Industry

Russian Federation, 111250, Russia, Moscow, Krasnokazarmennaya street, 14, building 1

Marina Nikolaevna Myznikova

National Research University "Moscow Power Engineering Institute"

Email: MyznikovaMN@mpei.ru

Candidate of Economic Sciences, Associate Professor, Associate Professor of the Department of Management in Energy and Industry

Russian Federation, 111250, Russia, Moscow, Krasnokazarmennaya street, 14, building 1

References

Supplementary files