Integrating Sustainable Development Goals into Renewable Energy Monopoly: A Generative AI Approach to Sustainable Development Education ^†

Abstract

This research aims to develop an educational board game, “Sustainable Home: Energy Challenge,” based on Monopoly by integrating sustainable development goals and renewable energy to use ChatGPT in human–computer collaboration. ChatGPT was used for game conceptualization, rule development, board creation, card design, and simulation in an iterative design. The developed board game demonstrated ChatGPT’s efficiency in educational game design and the benefits of human–computer collaboration. Game simulations validated the board game’s potential as a simulation tool to enhance diversity, cooperation, and strategic depth. The game effectively promoted SDG engagement and sustainable development education in gamified learning.

1. Introduction

Tabletop games, commonly known as board games, are a form of unplugged entertainment that relies on physical components such as boards, cards, and dice for interaction. With a long history and diverse genres, ranging from highly strategic war games to cooperative games emphasizing collaboration and relaxed party games, board games have become essential tools for leisure, social interaction, and education due to their rich gameplay and interactivity [1]. Board games offer entertainment experiences and promote cognitive development, social skill cultivation, and problem-solving ability enhancement [2].

Monopoly stands out as one of the most representative and influential classics among numerous board games. Invented by Charles Darrow in 1933, Monopoly was inspired by an earlier property game, “The Landlord’s Game.” With its simple and accessible rules, rich strategic depth, and mechanisms simulating real estate transactions, Monopoly has gained global popularity, becoming one of the best-selling commercial board games worldwide [3]. Monopoly’s success lies in its entertainment value and its effective integration of complex economic concepts into simple and understandable game rules, implicitly enabling players to learn financial and business knowledge during gameplay [4].

The educational value of Monopoly extends beyond learning economic knowledge to fostering strategic thinking and decision-making skills. O’Neill & Holmes [5] pointed out that Monopoly effectively trains players’ strategic thinking, negotiation, and social skills. In the game, players need to constantly analyze situations, assess risks, make decisions, and compete with other players, enhancing players’ cognitive and social abilities. Furthermore, Monopoly has also been applied in education as a supplementary tool for economics teaching, helping students understand market mechanisms and economic operations more intuitively [6].

However, the themes and values of traditional Monopoly games appear outdated today. The game’s pursuit of maximizing personal wealth and even causing other players to go bankrupt conflicts with societal advocacy for sustainable development and win–win cooperation. As sustainable development continues to attract attention, board games need to be explored in terms of how the Monopoly framework incorporates societal contemporary problems to create educational and socially valuable games. In this study, the Monopoly game framework, combined with the sustainable development goals (SDGs) and renewable energy themes, was used to design an educational board game, “Sustainable Home: Energy Challenge.”

Recently, artificial intelligence (AI) technology has undergone rapid development, particularly with regard to large language models (LLMs), affecting education. LLMs are natural language processing models based on deep learning technology which are capable of understanding and generating human language and performing complex language tasks such as text generation, question answering, translation, and dialogue [7]. LLMs exhibit performances exceeding those of previous models in natural language processing tasks, opening up new possibilities for human–computer collaboration and automated content generation [8].

ChatGPT 4o, developed by OpenAI, is a conversational AI based on an LLM. With its exceptional language comprehension ability, fluent and natural dialogue interaction, and rich knowledge base, it has garnered widespread attention. ChatGPT engages in natural language conversations with users and assists users in creative writing, code generation, information retrieval, and problem-solving tasks. In education, the potential of ChatGPT has attracted widespread attention. Kasneci et al. [9] suggested that ChatGPT could be used as a personalized learning assistant for students, providing instant academic guidance and customized learning resources. Mai et al. [10] pointed out that ChatGPT assists teachers in instructional design and content creation, reducing teachers’ workload and improving the quality of their teaching.

In educational design, human–machine collaboration is important. Liu et al. [11] proposed that educational designers need to combine human creativity and critical thinking with AI’s efficient content generation and knowledge integration to more efficiently develop innovative educational products through human–computer collaboration. ChatGPT, as a powerful language model, also shows potential for application in board game design. In this study, ChatGPT was used as a collaborative partner to design and simulate the “Sustainable Home: Energy Challenge” and validate the application value of LLMs in educational game design.

The SDGs, which were adopted by the United Nations in 2015, are a global action plan to achieve sustainable development in economic, social, and environmental dimensions worldwide by 2030 [12]. The SDGs encompass 17 goals, guiding global joint efforts to address urgent challenges such as poverty, hunger, climate change, inequality, and environmental degradation. Among these, SDG 7, “Affordable and Clean Energy,” and SDG 13, “Climate Action”, are directly related to renewable energy issues, highlighting the urgency and importance of energy transition and climate action.

Faced with challenges such as global climate change, the need to develop renewable energy sources has become a worldwide consensus and is becoming increasingly urgent. The International Renewable Energy Agency [13] stated that renewable energy sources such as solar, wind, hydro, geothermal, and bioenergy are key to achieving energy independence, environmental sustainability, and economic development. However, renewable energy development and widespread adoption require technological innovation and policy support to enhance public awareness and participation. Public understanding and support for renewable energy directly contribute to promoting renewable energy policies and the progress of energy transition [14].

Education plays a crucial role in promoting sustainable development and renewable energy issues. The United Nations Educational, Scientific, and Cultural Organization [15] emphasized that education is essential for achieving sustainable development goals. Education for sustainable development (ESD) aims to provide citizens with sustainable development literacy, enabling them to understand the complexity of sustainable development and promoting the sustainable development of society, the economy, and the environment. As an edutainment learning tool, board games provide an effective way to promote sustainable development education [16]. By gamifying sustainable development issues and integrating them into game mechanics, board games effectively enhance learners’ participation and learning motivation and promote the popularization of sustainable development knowledge and the establishment of values.

This research aims to explore the innovative design and simulation of the sustainable development educational board game “Sustainable Home: Energy Challenge” using ChatGPT. Relevant literature on board game design, Monopoly, LLMs, ChatGPT, SDGs and renewable energy, and agent-based modeling (ABM) was reviewed to establish the research’s theoretical foundation. Subsequently, the human–computer collaborative workflow was established using ChatGPT for board game design, and key design outcomes, including game concepts, rules, board, characters, cards, and SDGs, were identified. ChatGPT was used for game simulation. The simulation results were analyzed to enhance game balance and educational significance based on data-driven optimization strategies. The board game’s design concepts, game mechanics, and simulation results were analyzed from the perspective of ABM. The potential of board games as complex system simulation tools was also assessed. The findings and contributions of this research are described to provide academic references for future research based on educational technology, artificial intelligence, and sustainable development education.

2. Human–AI Collaborative Workflow Using ChatGPT

In this research, human designers’ creativity and strategic thinking were combined with the language generation and knowledge integration capabilities of the LLM as human–computer collaboration. ChatGPT was used to design the board game “Sustainable Home: Energy Challenge.” This collaborative model was refined in an iterative cycle. Designers used prompts to define design requirements, and ChatGPT responded with suggestions. Through multiple rounds of dialogue and optimization, the various aspects of the board game design were refined. The workflow was summarized into nine stages, covering the entire process from game concept establishment to detailed rule design.

In the concept establishment stage, designers described the board game design goals and framework to ChatGPT. They stated that the game should be based on Monopoly, and should incorporate the SDGs and renewable energy issues. ChatGPT established the game “Sustainable Home: Energy Challenge” and outlined its core design concept, laying the foundation for subsequent design directions. Next, in the game rule and board design stage, designers asked ChatGPT to design a 40-square game board. Based on the Monopoly structure and design concept, ChatGPT produced a preliminary board design, including corner squares, energy facility squares, and event squares, showcasing the framework of the game. In the game board content refinement stage, designers commanded ChatGPT to design player characters and assets to enrich game content and strategies. ChatGPT designed six characters with different professional backgrounds and characteristics and detailed how their character traits affect the game, increasing its role-playing elements and strategic depth (Figure 1).

The game’s rules were designed considering the completeness and playability of the regulations. Designers asked ChatGPT to describe the rules of the game. ChatGPT elaborated on the game’s objectives, processes, asset management, trading, card usage, and other details related to the game’s basic playability. To increase the fun and randomness of the game, opportunity and environmental challenge cards were created. ChatGPT generated various cards describing positive effects (such as subsidies and technology upgrades) and adverse effects (such as facility damage and reduced income), enriching the game’s event elements and strategic options. Regarding asset and fund management, the rules were formulated according to the game’s economic mechanism. Designers asked ChatGPT to detail the rules and gameplay of asset and fund management. ChatGPT elaborated on the rules of asset purchase, upgrade, trading, auction, disposal, income, fund sources, expenditures, loans, and repayments, making the game’s economic system more complete. To ensure the educational significance of the game, SDGs designs were integrated by ChatGPT, which detailed the concepts, design, completion conditions, and settlement methods. By integrating the SDGs into the game, the sustainable development educational connotation was enhanced.

The human–computer collaborative workflow is summarized as follows.

• Phase 1: Concept establishment—establish a game theme and core concept.
• Phase 2: Game rule and board design—design the basic structure of the game board.
• Phase 3: Game board content refinement—design player characters and assets.
• Phase 4: Game rule detail design—refine basic game rules and processes.
• Phase 5: Sustainable opportunity card design—design positive effect cards.
• Phase 6: Environmental challenge card design—design negative effect cards.
• Phase 7: Asset management rule and gameplay refinement—design an asset management mechanism.
• Phase 8: Fund management rule and gameplay refinement—design a fund management mechanism.
• Phase 9: SDGs-based rule design and gameplay refinement—integrate the SDGs and the reward mechanism.

In the human–computer collaboration, ChatGPT played a crucial role as a partner in each stage of game development. ChatGPT generated rich game content based on designers’ prompts and provided concrete and feasible design solutions and rule suggestions, significantly improving the efficiency and quality of board game design. The application of human–computer collaboration demonstrated the potential of LLMs in educational game design, providing a novel mode and direction for future educational game development.

3. Board Game Simulation and Optimization

To validate the design scheme of the developed board game and optimize it based on empirical data, ChatGPT was used to conduct multiple rounds of game simulations.

3.1. Game Simulation Design and Settings

Game simulation aims to replicate actual board game-playing and quantitatively analyze the impact of game parameters and rule designs on game dynamics. In the developed game, multiple key variables were controlled to ensure the validity and comparability of the simulation results.

Considering board games’ interactivity and gameplay experience, initial simulations were conducted by four players. Later, the number was adjusted to three players to focus more on the mechanisms and simplify the analysis. Different numbers of players were used to observe the game’s dynamics in different social scenarios. The initial game time was set at 2 h. In later stages, the game time was shortened to 1 h to assess the impact of different game durations on players’ strategy choices and game strategies. The initial capital was set at 3000 game currency units and later increased to 5000. By adjusting the initial capital, its impact on game-opening strategies, economic development speed, and player risk tolerance was monitored.

In each simulation, different player characters were employed to evaluate the roles and balance of different character traits in the game. The main character combinations were as follows.

Combinations including all character types: Player A (Wind Energy Pioneer), Player B (Policy Advocate), Player C (Solar Energy Technologist), and Player D (Energy Storage Expert).

Simplified character-type combinations: Player A (Wind Energy Pioneer), Player C (Solar Energy Technologist), and Player B (Policy Advocate).

The simulation was conducted in 18–22 rounds, and later shortened to 10 rounds. Simulations involving different numbers of rounds were conducted to observe the game’s development trends at different stages and the differences between long-term strategies and short-term strategies. In the simulation, the researcher played the bank and environmental character and executed the rules of the game to respond to players and record game data. Player actions, asset changes, and SDG target achievement were recorded in each round for subsequent analysis and optimization.

3.2. Result of Game Simulation

3.2.1. Validation and Strategy Divergence (Four Players, 2 h, and Initial Capital 3000)

In simulation, the feasibility of game rules and processes was validated to observe players’ strategy choices and game situation evolution. As shown in

Table 1. Summary of simulation results.

Player Character	Final Capital	Total Asset Value	SDG Score	Total Score
Player A (Wind energy pioneer)	1850	490	35	2375
Player B (Policy advocate)	1950	410	25	2385
Player C (Solar energy technologist)	1900	460	45	2405
Player D (Energy storage expert)	1725	470	30	2225

, different players exhibited significant strategy divergence based on their characters and understanding of game mechanisms.

Player C had the highest total score, followed closely by Player B, while Players A and D lagged behind.

Strategy divergence and character traits: Players’ strategy choices revealed their character traits. Player A focused on purchasing and upgrading wind energy facilities, reflecting their character trait of a “10% discount on wind energy facility purchase, 5% increase in income.” Player C actively developed solar energy and energy storage facilities and effectively utilized their character’s advantage of a “10% discount on solar power plant purchase, 20% reduction in upgrade cost.” Player B invested evenly in different types of energy facilities and actively used “Sustainable Opportunity” cards, demonstrating the diverse application of their character traits in strategies. Player D focused on upgrading energy storage facilities to increase their long-term income, which is consistent with their character trait of “10% discount on energy storage facility purchase, 10% increase in income.”

Significant differences in SDG achievement: Significant differences were observed in the degree of SDG achievement among different players. Player C had a significantly higher SDG score, mainly due to their active participation in cooperative transactions, effectively promoting SDG 17 (Partnerships for the Goals). The SDG scores of other players were relatively concentrated on specific goals. For example, Player A mainly achieved SDG 13 (Climate Action) while Player D achieved SDG 11 (Sustainable Cities and Communities). This indicated that the game’s rules had room for improvement in encouraging players to pursue diverse SDGs.

Game time and initial capital need adjustment: The simulation results showed that the game time was too long, lasting close to 2 h, which affected players’ game experience. In addition, the initial capital of 3000 game currency units appeared insufficient, limiting players’ development speed in the game’s early stages, possibly leading to a slower game rhythm.

3.2.2. Adjusting Initial Capital and Game Time (Four Players, 1 h, and Initial Capital 5000)

Based on the initial simulation results, the initial capital and game time were adjusted to enhance the game rhythm and players’ experience. The adjusted game showed an improved rhythm and experience and offered new challenges. After adjusting the initial capital and game time, the total score differences between players narrowed, but the overall ranking was the same as in the initial simulation results as shown in

Table 2. Summary of simulation results after adjusting the initial capital and game time.

Player Character	Final Capital	Total Asset Value	SDG Score	Total Score
Player A	1850	490	20	2360
Player B	1950	410	20	2380
Player C	1900	460	35	2395
Player D	1725	470	20	2215

.

Accelerated game rhythm: After increasing the initial capital to 5000 game currency units, players purchased and upgraded facilities more quickly in the game’s early stages. The game’s rhythm significantly accelerated, and the game was completed within 1 h, meeting the expected playtime of general board games.

Conservative player strategy: After shortening the game time, players’ strategy choices became conservative. They preferred to establish facilities with stable income quickly and were less inclined to engage in high-risk, high-return investments or cooperation. This was attributed to the limited game time, which precluded long-term strategic planning.

Reduced SDG achievement: With the shortening of game time, the time available for the players to achieve the SDGs was reduced, and the scores decreased. Pursuing game rhythm and playtime undermined the educational significance of the game, requiring a balance to be sought in subsequent optimizations.

3.2.3. Rule Optimization (Three Players, 1 h, and Initial Capital 5000)

To enhance the attractiveness and diversity of the SDGs, the game rules based on user-generated progressive suggestions were adjusted (as shown in

Table 3. Summary of simulation results with rule optimization.

Player Character	Final Capital	Total Asset Value	SDG Score	Total Score
Player A	4404	780	50	5184
Player B	4060	1020	100	5080
Player C	4015	940	50	4955

), which included increasing multi-goal reward mechanisms, goal progress-tracking mechanisms, facility type diversification requirements, promoting trading behavior, phased SDG rewards, and dynamic goal configuration. The rule optimization significantly promoted players’ diverse SDG achievement, as shown in

Table 4. Comparison of SDG achievement rate before and after optimization.

SDG	Achievement Rate
	Before Optimization (%)	After Optimization (%)
SDG 7 (Affordable and Clean Energy)	25%	100%
SDG 9 (Industry, Innovation and Infrastructure)	0%	33%
SDG 11 (Sustainable Cities and Communities)	25%	33%
SDG 13 (Climate Action)	25%	33%
SDG 17 (Partnerships for the Goals)	25%	67%
Average rate	20%	53.2%

.

After the rule optimization, the total scores of all players significantly improved, and Player A surpassed Player B, becoming the player with the highest total score. The average SDG achievement rate significantly increased from 20% before optimization to 53.2% Table 3 and Table 4).

The SDG 7 achievement rate significantly increased after rule optimization. All players achieved SDG 7 (Affordable and Clean Energy). The multi-goal reward mechanism and facility type diversification requirements effectively enhanced players’ focus on SDG 7.

The diverse goal achievement rate was improved. Players no longer focused on a single goal when achieving the SDGs, but achieved multiple SDGs. For example, Player B, after rule optimization, simultaneously achieved SDG 7, SDG 9, SDG 11, SDG 13, SDG 17, and other goals, demonstrating the incentive effect of the multi-goal reward mechanism.

Cooperative behavior became more prevalent. To achieve SDG 17 (Partnerships for the Goals), cooperative behavior among players significantly increased. For example, Players B and C conducted facility transactions and joint investments multiple times during the game, jointly promoting the phased goals of SDG 17 and validating the rule optimization effect of promoting transaction behavior.

The optimized game rules significantly enhanced the achievement of SDGs, diverse goal achievement, and the appeal of cooperative mechanisms. The strategic depth and educational significance of the game were improved.

3.3. Impact of Game Optimization on Balance and Educational Significance

Based on multiple game simulations and data analysis, the game optimization conducted positively impacted the balance and educational significance of the “Sustainable Home: Energy Challenge” board game. Rule optimization, while improving the attractiveness of SDGs, impacted game balance. Simulation results after optimization showed that the total score differences between players increased. Players’ strategy choices became diverse, and different strategy paths enabled advantages. The game balance adjustment was observed and fine-tuned in subsequent iterations. The optimization significantly improved the average SDG achievement rate, encouraging players to pursue SDGs actively and acquire sustainable development knowledge. Multi-goal reward mechanisms, facility type diversification requirements, and rule optimization promoted transaction behavior to effectively embody the comprehensiveness and complementarity of SDGs, enhancing the game’s educational value.

The game simulation and rule optimization conducted through ChatGPT validated the design of the “Sustainable Home: Energy Challenge” board game and significantly enhanced its balance and educational significance. The optimized game rules effectively balanced entertainment and educational value, laying a foundation for game promotion and application.

4. Conclusions

Using ChatGPT, a human–computer collaborative innovation model was established to design and simulate the board game “Sustainable Home: Energy Challenge” developed for the purpose of sustainable development education. ChatGPT’s potential in educational game design was validated from the ABM perspective during board game design and optimization.

Through the human–computer collaboration, educational designers can effectively utilize ChatGPT’s understanding, knowledge integration, and creative ideation to improve the efficiency and quality of board game design. ChatGPT demonstrated excellent assistance capabilities in game concept establishment, rule design, board generation, card design, and game simulation stages, providing a novel tool and method for educational game design. The developed board game successfully integrated SDGs and renewable energy themes into its mechanism, enabling players to experience the challenges and fun of energy company operation during the game, learn sustainable development knowledge, and cultivate sustainable development literacy. The optimized game rules effectively enhanced the attractiveness and diversity of SDGs, strengthened cooperative interaction among players, and improved the strategic depth and complexity of the game.

The game mechanics and simulation results were analyzed in ABM to validate the potential of the board game as a system simulation tool. The theoretical framework of ABM helps us to understand the behavior patterns of players, environmental factors, and interaction relationships in the board game system, and how they shape the emergent patterns at the system level. Game optimization strategies based on the ABM perspective effectively enhanced the game’s balance, educational significance, and strategic depth, providing a systematic methodology for board game design and optimization.

The result of our game result analysis provides empirical evidence supporting ABM’s application in sustainable development education. The board game, combined with ABM’s complex system simulation capability, helps learners intuitively understand the system complexity and emergent properties of sustainable development issues, cultivating systemic thinking and interdisciplinary integration capabilities.

While ChatGPT was used as a game simulation tool in this study and performed excellently in terms of language understanding and text generation, it required complex system simulation and presented limitations with regard to data analysis. Therefore, it is necessary to conduct further research using professional ABM simulation software such as NetLogo 6.4.0 or MASON 22 to conduct precise simulations for data analysis. The developed board game simplified the complexity of the real world. For example, energy markets, policy environments, technological development, and other aspects of the game were simplified, failing to reflect the complexity and uncertainty of the real market. The complexity of the game model needs to be added by incorporating diverse environmental factors and market mechanisms to enhance the game’s simulation realism and educational depth. This research was conducted for game design and simulation using ChatGPT, and lacked validation through user testing. Therefore, it is required to conduct user testing by inviting players with different backgrounds and age groups and collecting their feedback, effectiveness evaluation, and game balance opinions to optimize game design. Game rhythm and playtime decrease the average SDG achievement rate. Therefore, the trade-off relationship between game time, game difficulty, and SDG achievement rate needs to be explored to see a better balance point to enhance game entertainment and educational significance.

Based on the limitations, further research needs to be conducted in the following directions.

This research was based on renewable energy themes and SDG 7 and SDG 13. Other sustainable development goals need to be included, including food security, water resource management, and biodiversity conservation. To enhance game promotion and accessibility, digital board game platforms need to be digitized and combine the functions of online game platforms to expand the game’s playing field and community interaction. ABM perspective and human–computer collaboration were applied for the sustainable development of educational board games. The design and development of the developed game can be applied to other games, such as history-focused educational games, civic education games, and science education games. To achieve this, the application of ABM in different educational game designs must be explored.

This “Sustainable Home: Energy Challenge” board game and the playing results demonstrated the potential of LLMs in educational innovation design and highlighted the value of board games as sustainable development educational tools. Through the human–computer collaboration in the ABM framework, a sustainable development educational board game was designed to be entertaining, educational, and strategic. Through simulation and iterative optimization, the quality and effectiveness of the game can be improved. The process of game development provides a reference for educational technology using AI. Sustainable development education can be enhanced through educational innovation and gamified learning. With the continuous development of AI technology and the increasing importance of sustainable development issues, educational game design based on human–computer collaboration and ABM is expected to promote sustainable development education and facilitate sustainable social transformation.