The resilient city model is a concept of urban planning and management aimed at improving the adaptability and resilience of cities in the face of various pressures and impacts. These pressures and impacts may include natural disasters (such as earthquakes, floods and hurricanes), climate change, human accidents (such as terrorist attacks and industrial accidents) and socio-economic changes. The resilient city model emphasizes the sustainability and resilience of a city to ensure that it can maintain its function and operation in the face of pressure and can quickly return to normal state (Meerow et al., 2019; Sara and Joshua, 2019).

This model involves multiple aspects of planning and management, including the following key elements: multi-functional infrastructure: The resilient city model advocates the construction of multi-functional infrastructure, which can not only meet daily needs but also perform other functions in emergency situations. For example, a building can be used as an office building or a school, but it can also serve as a shelter in the event of a disaster; multi source energy supply: The resilient city model encourages cities to adopt multi source energy supply to reduce dependence on a single energy source. This includes utilizing renewable energy and energy reserve systems to ensure that cities can maintain operations in the event of energy supply disruptions; community participation and cooperation: The resilient city model emphasizes community participation and cooperation, encouraging residents to participate in urban planning and decision-making processes. This can enhance the cohesion and unity of the community, making the community work together to respond and recover in the face of crises; data-driven decision-making: The resilient city model advocates the use of data and information technology to support decision-making. By collecting, analyzing and utilizing urban data, it is possible to better understand the vulnerability and potential risks of cities and take corresponding measures to enhance their resilience (Moreno et al., 2021); disaster risk management: The resilient city model emphasizes the importance of disaster risk management in disaster risk management. This includes measures such as conducting risk assessments, establishing emergency response mechanisms, improving disaster warning systems and strengthening the seismic capacity of buildings and infrastructure to reduce losses caused by disasters.

The resilient city model is a comprehensive concept aimed at establishing sustainable, resilient and adaptable cities. By comprehensively considering factors such as urban planning, infrastructure construction, community participation and disaster management, the resilient city model can enhance the city’s response capacity, reduce risks and improve the quality of life of residents.

Artificial intelligence plays an important role in resilient city models. Artificial intelligence technology can provide support for urban planning, management and disaster response, improving the resilience and adaptability of cities. The application of artificial intelligence in resilient city models includes data analysis and prediction. Artificial intelligence can analyze large-scale urban data, including meteorological data, traffic data, population data, etc., to predict potential disasters and risks. Through machine learning and data mining techniques, potential threats faced by cities can be identified. Preventive and disaster reduction measures can be taken in advance. For smart energy management, artificial intelligence can be used for urban energy management to improve energy efficiency and sustainability by optimizing energy allocation and consumption. Technologies such as smart grid management, energy demand forecasting and distributed energy systems can provide flexibility and resilience to respond to changes in energy supply and emergencies. With the most important aspect of intelligent disaster management, artificial intelligence can play a crucial role in disaster management. Through intelligent sensor networks and data analysis, natural disasters such as earthquakes, floods and storms can be monitored and warned in real-time. Technologies such as intelligent emergency response systems and intelligent rescue robots can improve the efficiency and accuracy of disaster response. In urban planning and design, artificial intelligence can support the process of urban planning and design, providing multiple solutions and decision support through simulation and optimization algorithms. Intelligent city modeling can evaluate the resilience and feasibility of urban planning schemes, helping decision-makers make better decisions. The artificial intelligence resilient city model is shown in Figure 1.

The use of artificial intelligence in resilient city models can help improve the responsiveness of cities, reduce risks and improve the quality of life of residents. However, attention also needs to be paid to the ethical and privacy issues of artificial intelligence to ensure that the technology is used ethically and legally.

The purpose of urban shelter planning is to ensure the life safety and wellbeing of urban residents in the event of a disaster, emergency or other dangerous event. Therefore, the theoretical basis of urban shelter planning mainly involves: disaster management theory, spatial planning theory, sociological and psychological theory, engineering and architectural theory, as shown in Table 1.

In summary, the theoretical foundation of urban shelter planning encompasses theoretical knowledge from multiple disciplines such as disaster management, spatial planning, sociology and psychology, engineering and architecture. These theories provide guidance for planners to develop appropriate shelter planning strategies to respond to various disasters and emergencies faced by cities.

The development of urban shelter planning is one of the important issues in the fields of disaster management and urban planning. With the impact of factors such as climate change, frequent natural disasters and population growth, the planning of urban shelters is receiving increasing attention. The planning of urban shelters is developing towards multi-functionality, sustainability, community participation, technological innovation and cross departmental cooperation, as shown in Table 2.

In the planning of shelters in oil and gas energy resource-based cities, many factors need to be considered. The primary principle of shelters is to ensure the safety of residents. This includes selecting safe locations, avoiding potential natural disasters and industrial accident risks, as well as adopting seismic, fire prevention and other related facilities and measures. The shelter should also have sufficient capacity to accommodate the residents and staff of the city. Meanwhile, it is necessary to consider the needs of different groups, such as the elderly, children and people with disabilities and provide facilities and services that meet various needs (Su et al., 2023). Secondly, the planning of shelters should consider sustainability principles, including energy conservation, effective resource utilization and environmental protection. It adopts measures such as renewable energy, water resource management and waste disposal to reduce their impact on the environment. Basic living facilities and services, such as food, water, sanitation, medical services and emergency communication systems, should also be considered to provide psychological support, education and entertainment services to meet the needs of residents.

The site selection of shelters should also take accessibility and traffic connectivity into consideration and choose locations close to residential areas, transportation hubs and emergency service facilities to ensure that residents can quickly reach the shelters and provide necessary transportation tools and route planning. Shelters should be closely integrated with disaster management and emergency response mechanisms, cooperating with local governments, rescue agencies and relevant departments to establish emergency evacuation plans, warning systems and emergency communication networks to ensure timely and effective response to disasters and emergencies. The planning process of shelters should actively promote community participation and education. It is advisable to encourage residents to participate in the planning process, provide relevant training and drills, enhance residents’ awareness and abilities, enable them to understand the importance of shelters and correctly utilize and protect these places in emergency situations.

These factors provide guidance for the planning of shelters in oil and gas energy resource-based cities and help ensure that cities can provide safe shelter and support the needs of residents in the face of emergencies and disasters. In the planning process, it is necessary to comprehensively consider the local special environment and needs and collaborate with relevant stakeholders to develop specific planning plans that are suitable for the local situation.

The spatial planning system of shelters in oil and gas energy resource-based cities needs to consider the following aspects: 1) distribution of shelters: It should be distributed in various areas of the city to cover the entire population. The location of the shelter should take into account residents in densely populated areas, industrial areas and important facilities to ensure that they can quickly reach the shelter; 2) location selection: It requires consideration of geographical conditions, geological conditions and potential natural disaster risks when choosing shelter. Shelter should be kept away from earthquake fault zones, flood areas and other potential hazardous areas. Besides, its location should be convenient for transportation and material supply; 3) capacity and functionality: The shelter should have sufficient capacity to accommodate the scale of urban residents. Based on the size and population density of the city, it is necessary to determine the size and number of shelters. Besides, shelters should also have basic living facilities, such as drinking water facilities, sanitation facilities, medical facilities and emergency communication equipment; 4) building structure and safety facilities: The building structure of the shelter should have safety performance such as earthquake resistance, fire prevention and explosion prevention. Buildings with sturdy structures and high durability should be selected as shelters. Besides, shelter should be equipped with safety facilities such as emergency lighting, firefighting equipment, emergency water supply and power supply equipment. 5) multi-functionality and sustainability: The planning of shelters should consider their multi-functionality in non emergency situations, which can be used as community centers, schools or cultural activity venues in daily life to meet the diverse needs of residents. In emergency situations, they can quickly transform into shelters to provide protection. Besides, it should focus on sustainability, including energy conservation, water resource management and waste disposal; 6) auxiliary facilities and transportation connections: The planning of shelters should also take into account the needs of auxiliary facilities and transportation connections. This includes supporting facilities such as parking lots, roads, pedestrian and bicycle lanes, as well as connections to the public transportation system to facilitate residents’ access to shelters.

In summary, the spatial planning system of shelters in oil and gas energy resource-based cities needs to comprehensively consider factors such as geography, population and safety and be combined with the overall planning and disaster management system of the city to provide safe and reliable shelters, ensuring the safety and protection of residents in emergency situations.

The oil and gas energy resource-based urban shelter planning model is a planning method that integrates urban characteristics, disaster risks and residents’ needs. The basic framework of the oil and gas energy resource-based urban shelter planning model is shown in Figure 2:

The above is the framework of a basic oil and gas energy resource-based urban shelter planning model, which needs to be adjusted and refined according to the characteristics and needs of the city for specific implementation. Meanwhile, sustainability and environmental considerations, as well as coordination with other urban planning and disaster management systems, also need to be taken into account.

Artificial bee colony algorithm is a heuristic optimization algorithm developed from the behavior of bee colonies. It simulates the behavior of bee populations in the process of searching for food, searching for the optimal solution through information exchange and collaboration. Artificial bee colony algorithm is widely used to solve various optimization problems (Jun-qing and Yun-qi, 2020; Nouria and Farhad, 2020). The basic principle of artificial bee colony algorithm is as follows:

Initializing bee colony: At the beginning of the algorithm, a random set of initial bee individuals (solution) is generated, called “worker bees”. Each worker bee represents a candidate solution: X i = x min + r a n d 1 . D x max − x min (1)

Among them, i = 1,2 , . . . , S N / 2 ; SN is the initial number of generated bee sources; X i is a D-dimensional vector.

Evaluating the fitness of bees: The fitness of each worker bee, that is, the quality of the solution, is evaluated according to the objective function of the problem (Yin et al., 2017; Shan et al., 2019).

Selecting greedy worker bees: According to the fitness value of worker bees, some workers with better fitness are selected as “greedy worker bees”. The greedy worker bees continue to search for nearby solution spaces. m , n are Randomly selected and m ≠ n , with: u i = x i h (2) u i m h + 1 = x i m + r a n d 1 , 1 x i m − x n m (3)

Following the search of worker bees: Each greedy worker bee searches in the nearby solution space according to certain rules to find a better solution. The search method can be local search, global search or random search: f i t i = 1 / 1 + f x , f x ≥ 0 1 + f x , f x < 0 (4) p i = 0.9 f i t i max f i t i + 0.1 (5) u i m h + 1 = x i m + r a n d 1 , 1 x max m − x min m (6)

Selecting the reconnaissance bee: Besides the greedy worker bee, the remaining worker bees are referred to as “reconnaissance bees”. They randomly select a new solution for search to explore other regions of the solution space (Deng et al., 2019; Zhao and Kok Foong, 2022).

Updating the location of worker bees: Based on search results, the location of worker bees (solution) is updated. If the new solution is better than the original solution, the location is update, otherwise keep the original position unchanged.

Repeating the search process: Steps 3 to 6 is repeated until the stop criteria are met, such as reaching the maximum number of iterations or finding a satisfactory solution: x i m h + 1 = x min m + r a n d 1 , 1 x max m − x min m (7)

Through multiple iterations and information exchange between worker bees, the artificial bee colony algorithm gradually converges to the optimal solution. It has better global search ability and faster convergence speed and is suitable for various types of optimization problems, including function optimization, combinatorial optimization, scheduling problems, etc. One advantage of artificial bee colony algorithm is its simplicity and ease of implementation, but when solving complex problems, parameter adjustments and problem modeling need to be tailored to specific situations to achieve better optimization results (Yu et al., 2022; Cao and Qian, 2023).

Artificial bee colony algorithm can be applied in the planning of shelters in oil and gas energy resource-based cities to optimize decision-making in terms of location selection and capacity planning. For example, optimizing the location of shelters by using artificial bee colony algorithm can find the optimal location of shelters considering multiple factors. The algorithm can consider factors such as disaster risk, resident distribution and transportation convenience and gradually optimize the location plan through the search behavior of individual bees to achieve the optimal results. Optimization of capacity planning for shelters with artificial bee colony algorithm can optimize the capacity planning of shelters. The algorithm can dynamically adjust the capacity of shelters based on population distribution and demand analysis to satisfy the shelter requirement in different disaster scenarios. Information exchange is achieved by allowing individual bees to search for different capacity plans and gradually optimize capacity planning results through information exchange and collaboration. In optimization of facility and resource allocation, artificial bee colony algorithm can be applied to optimize the configuration of facilities and resources. The algorithm can consider different types of facilities and resources, such as drinking water, food reserves, medical equipment, etc. and optimize their allocation and configuration through the search behavior of individual bees to meet the needs of shelters to the greatest extent possible. For the optimization of spatial layout and safety design, artificial bee colony algorithm can be applied to optimize the spatial layout and safety design of shelters. The algorithms can optimize the layout of different functional areas, such as residential areas, activity areas, medical areas, etc., to maximize space utilization and improve residents’ safety. Through the search and information exchange of individual bees, the layout plan is gradually optimized to achieve better results. In optimization of emergency response mechanism, artificial bee colony algorithm can be applied to optimize the emergency response mechanism of shelters. It can consider optimization in terms of open and closed mechanisms, personnel evacuation and transfer mechanisms, etc. By searching and collaborating with individual bees, the optimal response strategy can be found to improve the efficiency and accuracy of emergency response (Guo, 2022; Wang, 2022).

The application of artificial bee colony algorithm in the shelter planning in oil and gas energy resource-based cities can provide faster, more accurate and comprehensive solutions, help decision-makers make more reasonable decisions and improve the resilience and adaptability of shelters. However, the specific application methods and parameter settings need to be adjusted and optimized according to the actual situation.