Large-scale hotel operation and maintenance management system based on BIM and Internet of Things

Abstract

The invention relates to a large-scale hotel operation and maintenance management system based on BIM and Internet of Things. The method comprises the following steps: collecting space management data and comprehensive performance data under different business scenarios in a historical period; space management data features are extracted, and an artificial intelligence model is trained offline by taking the comprehensive performance data as corresponding labels; inputting the space management data of the current hotel into the trained artificial intelligence model, and outputting the comprehensive performance under the current business scene; and according to the comprehensive performance under the current business scene, continuously adjusting the space management of the hotel until a preset comprehensive performance threshold value is reached. The problem that the operation and maintenance management efficiency is low due to the fact that an existing hotel is difficult to adapt to changes of different business scenes is solved. By constructing the multi-layer perceptron model of the space management data and the comprehensive performance data, the comprehensive performance is predicted according to the current space management data, and the accuracy and effect of space management are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of building operation and maintenance management, and relates to a large-scale hotel operation and maintenance management system based on BIM and the Internet of Things. Background Art

[0002] BIM (Building Information Modeling) is an innovative tool based on three-dimensional digital technology. It integrates various information of large-scale hotel construction projects, including three-dimensional geometry, component materials, performance, price, location and other aspects. With its powerful three-dimensional visualization function, BIM provides managers with intuitive hotel layout and structure display, which facilitates the rapid location of equipment and facilities. At the same time, it also integrates rich information, realizes centralized information management and sharing, helps hotels to plan and manage space, optimize room layout and utilization efficiency, and monitor room occupancy in real time, playing an important role in improving the operational efficiency, service quality and customer satisfaction of large hotels.

[0003] However, for large hotels, there are often changes in various business scenarios (such as passenger flow, weather, seasonal changes, special events such as large conferences, weddings, etc.). The use of BIM for spatial planning and management of equipment, facilities, resources, etc. is difficult to adapt to changes in different business scenarios, resulting in inefficient operation and maintenance management. Summary of the invention

[0004] The present invention provides a large-scale hotel operation and maintenance management system based on BIM and the Internet of Things, which aims to collect BIM space management data and its corresponding performance data of different business scenarios of large hotels in historical periods through the Internet of Things technology, and construct an artificial intelligence model to analyze the relationship between space management data and performance data, so as to solve the problem that the existing technology is difficult to adapt to changes in different business scenarios, resulting in low efficiency of operation and maintenance management.

[0005] The present invention provides a large-scale hotel operation and maintenance management system based on BIM and the Internet of Things, comprising a scenario recognition module, a BIM module, a performance analysis module and a decision adjustment module, wherein the scenario recognition module, the BIM module, the performance analysis module and the decision adjustment module are communicatively connected, wherein: The scenario recognition module is used to recognize the current hotel's business scenario data; The BIM module is used to obtain the current hotel's space management data in real time through BIM; The performance analysis module is used to analyze the comprehensive performance of the hotel in the current business scenario based on the current hotel space management data, and includes the following steps: S1. Collect space management data and comprehensive performance data under different business scenarios in historical periods. The comprehensive performance data is comprehensively evaluated through the operational efficiency, service quality and customer satisfaction of large hotels; S2, extract the features of space management data, and use the comprehensive performance data as the corresponding labels to train the artificial intelligence model offline; S3: Input the current hotel's space management data into the trained artificial intelligence model and output the comprehensive performance under the current business scenario; The decision adjustment module is used to continuously adjust the hotel's space management according to the comprehensive performance under the current business scenario until a preset comprehensive performance threshold is reached.

[0006] Furthermore, the business scenario data includes customer flow conditions, weather changes, seasonal changes and special events; the space management data includes equipment distribution data, functional area division data, service personnel allocation data and safety measures management data.

[0007] Furthermore, the spatial management data is visualized using GIS.

[0008] Further, in step S1, the comprehensive evaluation of the operation efficiency, service quality and customer satisfaction of the large hotel includes the following steps: S11. Determine the evaluation index system: clarify the evaluation indexes of operation efficiency, service quality and customer satisfaction, among which: the evaluation indexes of operation efficiency include equipment utilization rate, staff work efficiency and energy consumption rate; the evaluation indexes of service quality include the integrity of facilities and equipment, the standardization and efficiency of service processes and the degree of personalization of services; the evaluation indexes of customer satisfaction include the guests' satisfaction evaluation of the overall environment of the hotel, service attitude, catering quality and accommodation comfort; S12. Quantitative evaluation: Quantify the evaluation indicators and obtain the comprehensive evaluation results of operational efficiency, service quality and customer satisfaction; S13. Determine weights: Determine the weights of operational efficiency, service quality and customer satisfaction based on the hotel’s actual situation and development priorities; S14. Comprehensive calculation: Based on the determined weights and quantified comprehensive evaluation results, a comprehensive calculation is performed to obtain the hotel's comprehensive performance score.

[0009] Furthermore, in step S2, extracting the space management data features includes the following steps: Numerical processing of spatial management data is performed to obtain multi-dimensional spatial management variable data; The principal component analysis method is used to reduce the dimension of multidimensional space management variable data to obtain a number of space management principal components, which are used as space management data features.

[0010] Furthermore, the spatial management principal component is used as the spatial management data feature, specifically by selecting the principal component with a total explained variance greater than 80%. nThe spatial management principal components are used as the spatial management data features.

[0011] Furthermore, in step S2, the artificial intelligence model is configured as a multi-layer perceptron, comprising the following steps: S21, data preparation: sort out the extracted spatial management data features and the corresponding comprehensive performance data labels, and divide the data samples into training set, validation set and test set; S22. Network structure design: determine the input layer, hidden layer and output layer neuron composition of the multi-layer perceptron; S23, Initialize weights and biases: assign initial values ​​to each connection weight and bias in the network; S24, forward propagation: Input a sample data in the training set into the multi-layer perceptron, and perform forward propagation through the calculation of each layer. In each layer, the input and weight are linearly combined, and the bias is added. Then, the output of each layer is obtained through the activation function, and the prediction result of the output layer is finally obtained after passing through the input layer and the hidden layer in turn. S25. Calculate loss: Compare the prediction result of the output layer with the actual comprehensive performance data label and calculate the value of the loss function; S26, back propagation: According to the value of the loss function, the gradient of the weights and biases of each layer is calculated through the back propagation algorithm; S27, parameter update: use the optimization algorithm to update the weights and biases in the network according to the calculated gradient; S28, model training and adjustment: repeat steps S24 to S27, perform multiple iterations of training on the entire training set, and during the training process, regularly use the validation set to evaluate the performance of the model, adjust the model's hyperparameters based on the results of the validation set, and stop training when the performance of the model on the validation set no longer improves; S29, Model evaluation: Use the test set to perform a final evaluation of the trained model.

[0012] Furthermore, the decision adjustment module includes a monitoring module, a troubleshooting module, a parameter adjustment module and an execution module, wherein: The monitoring module is used to monitor and analyze whether the comprehensive performance under the current business scenario is lower than a preset comprehensive performance threshold; The troubleshooting module is used to troubleshoot the space management data when the comprehensive performance is lower than a preset comprehensive performance threshold, and determine abnormal space management data that does not conform to the current business scenario; The parameter adjustment module is used to adjust the abnormal space management data and re-input the artificial intelligence model until the output comprehensive performance is not lower than the preset comprehensive performance threshold, and the adjustment is completed; The execution module performs space management on the hotel according to the adjusted space management data.

[0013] Furthermore, when the comprehensive performance has not reached a preset comprehensive performance threshold, the decision adjustment module notifies the management personnel to perform manual space management on the hotel.

[0014] Beneficial effects of the present invention: (1) By collecting space management data and comprehensive performance data under different business scenarios in historical periods; extracting space management data features, and using comprehensive performance data as corresponding labels, offline training of artificial intelligence models; inputting the space management data of the current hotel into the trained artificial intelligence model, and outputting the comprehensive performance under the current business scenario; according to the comprehensive performance under the current business scenario, continuously adjusting the space management of the hotel until the preset comprehensive performance threshold is reached. The present invention solves the problem that existing hotels are difficult to adapt to changes in different business scenarios, resulting in low efficiency of operation and maintenance management.

[0015] (2) By constructing a multi-layer perceptron model of spatial management data and comprehensive performance data, it is possible to predict the comprehensive performance based on the current spatial management data, thereby improving the accuracy and effectiveness of spatial management. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0017] Figure 1 This is a structural diagram of a large-scale hotel operation and maintenance management system based on BIM and the Internet of Things in the present invention.

[0018] Figure 2 FIG. 4 is a flow chart of a performance analysis module in one embodiment of the present invention.

[0019] Figure 3 It is a structural diagram of a decision adjustment module in one embodiment of the present invention. DETAILED DESCRIPTION

[0020] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation mode, structure, characteristics and effects of the present invention are described in detail below in combination with the accompanying drawings and preferred embodiments.

[0021] See also Figure 1-Figure 3 The present application provides a large-scale hotel operation and maintenance management system based on BIM and the Internet of Things, including a scenario recognition module, a BIM module, a performance analysis module and a decision adjustment module, wherein the scenario recognition module, the BIM module, the performance analysis module and the decision adjustment module are communicatively connected, wherein: The scenario recognition module is used to identify the current business scenario data of the hotel, wherein the business scenario data includes passenger flow, weather changes, seasonal changes and special activities; In this embodiment, the scenario recognition module is used to accurately identify the current hotel's business scenario: in terms of passenger flow, the module can carefully distinguish the peak passenger flow state. When the passenger flow is at its peak, the hotel's lobby, restaurant, elevator and other public areas will be crowded. At this time, the scenario recognition module will monitor the substantial increase in the flow of people in each area, providing an important basis for the efficient operation of the hotel's service personnel and facilities and equipment. For example, in the lobby area, it may be necessary to increase the number of front desk staff to speed up the check-in and check-out of guests; in the restaurant, it may be necessary to prepare more ingredients in advance to meet the dining needs of a large number of guests. In the low passenger flow state, the various areas of the hotel are relatively deserted, and the scenario recognition module will promptly feedback this situation so that the hotel can reasonably adjust service resources and avoid resource waste. For example, the number of service personnel can be appropriately reduced to reduce energy consumption.

[0022] In terms of weather changes, the scenario recognition module will pay close attention to various meteorological conditions. On sunny days, the availability of outdoor facilities such as swimming pools and gardens is greatly increased. At this time, the module will remind hotel managers to strengthen the maintenance and management of outdoor facilities to ensure that guests can enjoy the outdoor environment safely and comfortably. At the same time, some outdoor activities and services can be reasonably arranged according to the sunny conditions, such as outdoor barbecues and yoga courses. On rainy days, the demand for indoor activities increases. The module will recognize this change and prompt the hotel to provide more indoor entertainment facilities and services, such as cinemas and indoor game rooms. In addition, rainy days may also affect the hotel's traffic and drainage systems. The scenario recognition module will issue an alarm in time so that the hotel can take appropriate measures to deal with it. Snowy days have specific effects on the hotel's traffic and heating. The module will monitor the snowy conditions and remind the hotel to strengthen the cleaning and anti-skid treatment of the roads to ensure the safety of guests' travel. At the same time, the hotel's heating system will be adjusted according to the temperature changes on snowy days to ensure a comfortable indoor temperature.

[0023] In terms of seasonal changes, different seasons will bring different service needs. In spring, everything comes back to life, and guests may have more outdoor activities. The scenario recognition module will recognize the characteristics of this season and recommend that the hotel open outdoor areas such as gardens and terraces to provide spring-themed catering and event services. In summer, the temperature rises and high requirements are placed on refrigeration equipment. The module will monitor the hot weather in summer to ensure the normal operation of the hotel's air-conditioning system, and can also provide some cool drinks and cooling activities. In autumn, the weather is clear and refreshing, and there may be special landscape appreciation needs. The module will arrange some outdoor autumn appreciation activities according to the characteristics of autumn, such as maple leaf viewing and orchard picking. In winter, the cold weather focuses on heating and indoor activities. The scenario recognition module will ensure the stable operation of the hotel's heating system, and provide some warm indoor activities, such as hot springs and indoor sports.

[0024] In terms of special events, including large meetings, the scenario recognition module will identify specific needs for conference rooms, audio equipment, projection equipment, etc. It will notify hotel managers in advance to arrange the conference room and debug the equipment to ensure the smooth progress of the meeting. At the same time, it will also reasonably arrange service personnel and catering services according to the scale and time of the meeting. In the wedding scene, the module will identify unique requirements for banquet hall layout, catering services, wedding planning, etc. It will assist hotel staff in creating a romantic and warm wedding scene, provide high-quality catering and services, and ensure the success of the wedding.

[0025] The BIM module is used to obtain the space management data of the current hotel in real time through BIM, and the space management data includes equipment distribution data, functional area division data, service personnel allocation data and safety measures management data; In this embodiment, BIM (Building Information Modeling) performs well in acquiring large-scale hotel space management data, where equipment distribution data, functional area division data, service personnel allocation data, and safety measures management data are not fixed in the hotel space, but can be flexibly allocated.

[0026] In terms of equipment distribution, the BIM model can clearly show the specific locations of various types of equipment in the hotel space. As the hotel's operating needs change, the distribution of equipment can be adjusted. For example, when the passenger flow in a certain area increases and the demand for equipment such as air conditioners increases, the equipment can be flexibly deployed according to the actual situation, and the equipment in some idle areas can be transferred to areas with higher demand to ensure the hotel's comfort and service quality.

[0027] Functional area division data is also flexible. Hotels can re-divide functional areas according to different business scenarios, such as special events, seasonal changes, etc. For example, when holding a large conference, some leisure areas can be temporarily transformed into conference auxiliary areas to meet the special needs of the conference. In the peak tourist season, the ratio of guest room area and dining area can also be adjusted according to passenger flow to better meet the needs of guests.

[0028] The flexible allocation of service personnel allocation data is particularly important. Service personnel can be dynamically deployed in the hotel space according to the changes in passenger flow and service demand in different areas. When the service pressure in a certain area increases, such as during peak hours in the restaurant, service personnel can be deployed from relatively idle areas in a timely manner to ensure efficient service for guests.

[0029] The data of safety measures management is not static in the hotel space. With the adjustment of hotel layout and the change of business scenarios, the distribution of safety facilities can also be optimized. For example, when holding large-scale events, temporary safety exit signs and monitoring equipment can be added to ensure the safety of the event. The flexible distribution of these data in the hotel space provides strong support for the hotel to adapt to various business scenario changes.

[0030] Furthermore, the spatial management data is visualized using GIS.

[0031] In this embodiment, GIS technology can present the hotel's spatial management data in an intuitive and clear way. For equipment distribution data, GIS can accurately mark the location of various types of equipment on the map, such as air conditioners, elevators, lighting facilities, etc. Managers can quickly understand the specific distribution of equipment through the visual interface. Once a device fails, they can quickly locate it and arrange maintenance personnel to handle it.

[0032] In terms of the visualization of functional area division data, GIS can use different colors or patterns to distinguish the various functional areas of the hotel, such as the guest room area, dining area, conference area, etc. In this way, whether hotel managers are allocating resources or providing guidance services to guests, they can use spatial information more efficiently.

[0033] The service personnel allocation data can be visualized in GIS to dynamically display the distribution of service personnel in different areas of the hotel. Through real-time updates, managers can adjust the service personnel's job arrangements in a timely manner according to factors such as changes in passenger flow to ensure the timeliness and efficiency of service.

[0034] For safety measures management data, GIS can clearly present the locations of safety facilities such as fire-fighting equipment and emergency exits, and can be combined with functions such as simulation drills to plan evacuation routes in emergency situations in advance, thereby improving the hotel's safety assurance capabilities.

[0035] The performance analysis module is used to analyze the comprehensive performance of the hotel in the current business scenario based on the current hotel space management data, and includes the following steps: S1. Collect space management data and comprehensive performance data under different business scenarios in historical periods. The comprehensive performance data is comprehensively evaluated through the operational efficiency, service quality and customer satisfaction of large hotels; Furthermore, the comprehensive evaluation of the operational efficiency, service quality and customer satisfaction of large hotels includes the following steps: S11. Determine the evaluation index system: First, we need to identify the specific evaluation indicators for operational efficiency, service quality, and customer satisfaction. For operational efficiency, we can include indicators such as equipment utilization, staff efficiency, and energy consumption rate. Equipment utilization can be determined by counting the ratio of the actual use time of the equipment to the available time; staff efficiency can be measured by indicators such as the average response time of service personnel and the amount of tasks completed per unit time; and energy consumption rate is the ratio of the total energy consumption of the hotel to the business volume.

[0036] The service quality can include indicators such as the integrity of facilities and equipment, the standardization and efficiency of service processes, and the degree of personalization of services. The integrity of facilities and equipment can be evaluated through regular inspections and maintenance records as well as customer feedback; the standardization and efficiency of service processes can be considered from aspects such as the degree of standardization of services and service time; and the degree of personalization of services can be judged based on how well the customer's needs for personalized services are met.

[0037] Customer satisfaction can include guest satisfaction evaluation indicators on the overall hotel environment, service attitude, catering quality, accommodation comfort, etc. Guest feedback information can be collected through questionnaire surveys, online review analysis, etc.

[0038] S12. Quantitative evaluation: The collected data is quantified for comprehensive evaluation. For operational efficiency indicators, the data can be converted into specific values ​​according to the set calculation formula. For example, the equipment utilization rate can be directly calculated to obtain a specific percentage value; the work efficiency of personnel can be quantified by setting corresponding scoring standards based on indicators such as response time and task volume.

[0039] Service quality indicators can also be quantified according to different indicator characteristics. The integrity of facilities and equipment can be divided into different levels and assigned corresponding scores according to the inspection results; the standardization and efficiency of service processes can be scored according to service time and standardization.

[0040] Customer satisfaction indicators can be quantified based on guests' evaluations, for example, very satisfied, satisfied, average, dissatisfied, and very dissatisfied can be assigned different scores, and then the average score can be calculated.

[0041] S13. Determine weight: According to the actual situation and development focus of the hotel, determine the weights of operational efficiency, service quality and customer satisfaction in the comprehensive evaluation. If the hotel focuses on cost control and resource utilization efficiency, it may give a higher weight to operational efficiency; if the hotel takes providing high-quality services as its core competitiveness, it may give a higher weight to service quality; if the hotel hopes to increase its market share and reputation, the weight of customer satisfaction may be relatively high.

[0042] S14. Comprehensive calculation: Based on the determined weights and the quantified evaluation index values, a comprehensive calculation is performed to obtain the hotel's comprehensive performance score. For example, the weighted average method can be used to multiply the quantified values ​​of operational efficiency, service quality, and customer satisfaction by the corresponding weights, and then add them together to obtain the comprehensive performance score.

[0043] S2, extract the features of space management data, and use the comprehensive performance data as the corresponding labels to train the artificial intelligence model offline; In this embodiment, extracting space management data features refers to finding key information that can reflect the characteristics of hotel space management from space management data such as equipment distribution data, functional area division data, service personnel allocation data, and safety measures management data. For example, features such as equipment type and location distribution law are extracted from equipment distribution data; features such as area ratio and connection relationship of different functional areas are extracted from functional area division data; features such as density of service personnel in different areas and job distribution characteristics are extracted from service personnel allocation data; and features such as types of safety facilities and location concentration are extracted from safety measures management data.

[0044] Then, using comprehensive performance data as labels means using comprehensive performance data obtained by evaluating the operational efficiency, service quality, and customer satisfaction of large hotels as classification labels for training artificial intelligence models. The comprehensive performance data is divided into several hierarchical categories, which serve as labels in classification learning.

[0045] Finally, the offline training AI model uses the large amount of space management data that has been collected and its corresponding comprehensive performance data to train the AI ​​model in an environment that is separated from the real-time business scenario. In this way, the model learns the relationship between the space management data characteristics and the comprehensive performance, so that in the future it can quickly predict the hotel's comprehensive performance based on new space management data, provide decision-making suggestions for optimizing space management, and thus improve the hotel's operational management level.

[0046] Furthermore, the extracting of space management data features comprises the following steps: Numerical processing of spatial management data is performed to obtain multi-dimensional spatial management variable data; In this embodiment, since the space management data involves multiple data types and cannot be directly used for modeling, in order to facilitate modeling analysis, the space management data is uniformly digitized. Specifically, the space management data is digitized to obtain multidimensional space management variable data, which can be performed according to the following steps: (1) Numericalization of equipment distribution data: Equipment quantity statistics: Count different types of equipment, such as the number of air-conditioning equipment, the number of elevators, the number of lighting equipment, etc., and directly convert them into specific values.

[0047] Quantification of device location: The location of the device in the hotel space is expressed in coordinates. For example, a fixed point in the hotel is used as the origin to establish a two-dimensional or three-dimensional coordinate system to coordinate the location of the device, thus converting the location information into a numerical value. Alternatively, the distance between the device and a specific reference point (such as the center of the hotel lobby) is calculated to obtain a distance value.

[0048] Equipment status assessment: The operating status of the equipment (such as normal operation, failure, and maintenance) can be coded and assigned a value. For example, the normal operation status is assigned a value of 1, the failure status is assigned a value of 0, etc.

[0049] (2) Numericalization of functional area division data: Calculation of functional area: The area data of each functional area is obtained by measurement or from architectural drawings and directly used as a numerical variable.

[0050] Quantification of functional area connectivity: The number of connection channels between functional areas can be counted, or the shortest path length between functional areas can be calculated as a numerical variable.

[0051] Functional area attribute coding: Specific attributes of functional areas, such as different room types in the guest room area (standard room, deluxe suite, etc.), can be coded. For example, the standard room is coded as 1, the deluxe suite is coded as 2, etc.

[0052] (3) Numericalization of service personnel allocation data: Statistics on the number of service personnel: directly count the number of service personnel in different areas as a numerical variable.

[0053] Service staff response time measurement: Record the service staff's response time to guest needs in seconds or minutes as a numerical variable.

[0054] Calculation of service personnel position distribution ratio: Calculate the ratio of service personnel in different positions to the total number of service personnel as a numerical variable.

[0055] (4) Digitalization of safety measures management data: Statistics of the number of safety facilities: The number of safety facilities such as fire-fighting equipment and surveillance cameras is counted as a numerical variable.

[0056] Quantification of safety facility inspection cycle: Quantify the inspection cycle of safety facilities in days.

[0057] The principal component analysis method is used to reduce the dimension of multidimensional space management variable data to obtain a number of space management principal components, which are used as space management data features.

[0058] In this embodiment, principal component analysis can effectively reduce the data dimension and extract the main features in the space management data processing, provide concise and effective data representation for subsequent data analysis and model building, and help to better understand and manage the hotel's space resources. Principal component analysis is a data dimensionality reduction technology that aims to extract the most important information from high-dimensional data and convert it into several principal components in low-dimensional space. First, for the multidimensional space management variable data obtained by numerical processing, principal component analysis finds the correlation structure in the data by calculating the covariance matrix of the data. The covariance matrix reflects the linear relationship between different variables, and high covariance means that there is a strong correlation between the variables. Then, the principal component direction of the data is determined by solving the eigenvalues ​​and eigenvectors of the covariance matrix. The larger the eigenvalue, the more information the corresponding principal component contains. When selecting the principal component, the number of principal components is usually determined based on the cumulative variance contribution rate. The cumulative variance contribution rate indicates the proportion of the total variance of the original data that the selected principal component can explain. Generally, the first few principal components whose cumulative variance contribution rate reaches a certain threshold (such as 80% or 90%) are selected, so that the dimension of the data can be greatly reduced while retaining most of the data information. The principal components are constructed by linearly combining the original variables. Each principal component is a weighted sum of the original variables, and the weight is determined by the components of the eigenvector. The principal component loading reflects the correlation between the original variables and the principal component. By analyzing the principal component loading, the actual meaning of each principal component can be explained.

[0059] Furthermore, the spatial management principal component is used as the spatial management data feature, specifically by selecting the principal component with a total explained variance greater than 80%. n The spatial management principal components are used as the spatial management data features.

[0060] The multilayer perceptron (MLP) is essentially similar to the traditional back propagation (BP) neural network. The multilayer perceptron is usually a feedforward neural network composed of multiple neurons, which can learn the mapping relationship between input data and output data by adjusting weights and biases. During the training process, an algorithm similar to back propagation is indeed used to calculate the gradient of the loss function to the parameters of each layer, and the parameters are updated through an optimization algorithm to minimize the loss function. It can be said that the multilayer perceptron is a generalized neural network structure, and the traditional BP neural network is one of the specific implementations. The multilayer perceptron can adopt different activation functions, optimization algorithms and network structure designs, and has certain flexibility and scalability. Although the multilayer perceptron is closely related to the BP neural network, the concept of the multilayer perceptron is broader and covers a variety of possible neural network structures and training methods.

[0061] Furthermore, the artificial intelligence model is configured as a multi-layer perceptron, comprising the following steps: S21. Data preparation: First, sort out the extracted spatial management data features and the corresponding comprehensive performance data labels, and divide them into training set, validation set, and test set. Usually, they can be divided according to a certain ratio (such as 70% training set, 15% validation set, and 15% test set). Ensure the quality and integrity of the data, and handle outliers and missing values.

[0062] S22. Network structure design: A multilayer perceptron usually consists of an input layer, a hidden layer, and an output layer. Determine the number of nodes in the input layer, that is, the dimension of the spatial management data features. According to the complexity of the problem and the characteristics of the data, select the appropriate number of hidden layers and the number of nodes in each hidden layer. Generally speaking, you can start with fewer hidden layers and nodes, and then adjust according to the performance of the model. The number of nodes in the output layer usually corresponds to the dimension of the comprehensive performance data. If it is a single comprehensive performance indicator (such as a numerical score), the output layer has only one node.

[0063] S23. Initialize weights and biases: Assign initial values ​​to each connection weight and bias in the network. Common initialization methods include random initialization and Xavier initialization. Appropriate initialization can help the model converge faster.

[0064] S24, forward propagation: Input a sample data in the training set into the multi-layer perceptron, and perform forward propagation through the calculation of each layer. In each layer, the input and weight are linearly combined, and the bias is added, and then the output of the layer is obtained through the activation function. Common activation functions include sigmoid, tanh, ReLU, etc. After passing through the input layer and hidden layer in turn, the prediction result of the output layer is finally obtained.

[0065] S25. Calculate loss: Compare the prediction results of the output layer with the actual comprehensive performance data label and calculate the value of the loss function. Commonly used loss functions include mean square error (MSE), cross entropy loss, etc. The loss function measures the difference between the model prediction and the actual value.

[0066] S26, Back Propagation: Based on the value of the loss function, the back propagation algorithm is used to calculate the gradient of the weights and biases of each layer. Starting from the output layer, the gradients of each hidden layer and input layer are calculated in turn. The back propagation algorithm uses the chain rule to pass the derivatives of the loss function to the parameters of each layer back layer by layer.

[0067] S27, parameter update: Use optimization algorithms (such as gradient descent, Adam optimizer, etc.) to update the weights and biases in the network according to the calculated gradients. The purpose of the update is to minimize the loss function so that the model's prediction results are closer to the actual value.

[0068] S28, model training and adjustment: Repeat steps S24 to S27 to perform multiple iterations of training on the entire training set. During the training process, the validation set is used regularly to evaluate the performance of the model, and the hyperparameters of the model (such as learning rate, number of hidden layer nodes, etc.) are adjusted according to the results of the validation set. When the performance of the model on the validation set no longer improves, training can be stopped to prevent overfitting.

[0069] S29, Model evaluation: Use the test set to perform a final evaluation on the trained model. Calculate the model's loss function value, accuracy and other indicators on the test set to evaluate the model's generalization ability. If the model performance is not ideal, you can readjust the model structure and parameters and train again.

[0070] S3: Input the current hotel's space management data into the trained artificial intelligence model and output the comprehensive performance under the current business scenario; In this embodiment, when the space management data of the current hotel is input into the trained artificial intelligence model, the model will process and analyze the new input data based on the relationship between the space management data characteristics and the comprehensive performance learned during the training process.

[0071] First, the pre-processed current space management data, including equipment distribution, functional area division, service personnel allocation, and safety measures management, will be formatted according to the model input requirements. These data have been digitized and feature extracted to be better understood by the model.

[0072] Then, after the data enters the model, it passes through the input layer, hidden layer, and output layer in sequence. In the input layer, the data is received and passed to the hidden layer. The neurons in the hidden layer perform weighted summation on the input data and perform nonlinear transformations through activation functions. Different hidden layers can learn different levels of feature representations, thereby capturing the complex relationships in spatial management data.

[0073] After being processed by multiple hidden layers, the data finally reaches the output layer. The output layer will output the comprehensive performance prediction results under the current business scenario based on the mapping relationship learned by the model during the training process. This result is a classification result, such as high, medium, and low comprehensive performance evaluation.

[0074] Finally, hotel managers can understand the hotel's operating status under the current business scenario based on the comprehensive performance results output by the model, and make corresponding decisions and measures accordingly. For example, if the comprehensive performance is low, managers can analyze which specific aspects of space management data have problems, and then adjust equipment distribution, optimize service personnel allocation, or strengthen safety measures management, etc., to improve the hotel's overall performance.

[0075] The decision adjustment module is used to continuously adjust the hotel's space management according to the comprehensive performance under the current business scenario until a preset comprehensive performance threshold is reached.

[0076] Furthermore, the decision adjustment module includes a monitoring module, a troubleshooting module, a parameter adjustment module and an execution module, wherein: The monitoring module is used to monitor and analyze whether the comprehensive performance under the current business scenario is lower than a preset comprehensive performance threshold; The troubleshooting module is used to troubleshoot the space management data when the comprehensive performance is lower than a preset comprehensive performance threshold, and determine abnormal space management data that does not conform to the current business scenario; The parameter adjustment module is used to adjust the abnormal space management data and re-input the artificial intelligence model until the output comprehensive performance is not lower than the preset comprehensive performance threshold, and the adjustment is completed; The execution module performs space management on the hotel according to the adjusted space management data.

[0077] In this embodiment, the decision adjustment module plays a key role in hotel operations. It continuously adjusts the hotel's space management based on the comprehensive performance under the current business scenario to ensure that the hotel's operations can reach the preset comprehensive performance threshold. This threshold represents the operating level that the hotel expects to achieve, covering multiple aspects such as operating efficiency, service quality, and customer satisfaction. By continuously adjusting space management, the hotel can better adapt to different business scenarios and improve its overall competitiveness.

[0078] Among them, the monitoring module continuously pays attention to the comprehensive performance under the current business scenario. It evaluates the comprehensive performance of the hotel by collecting and analyzing various data, such as equipment operation status, service staff work efficiency, customer feedback, etc. At the same time, the current comprehensive performance is compared with the preset comprehensive performance threshold to determine whether the hotel's operation has reached the expected level. If the comprehensive performance is lower than the threshold, the monitoring module will trigger the troubleshooting module for further analysis.

[0079] When the monitoring module finds that the comprehensive performance is lower than the preset comprehensive performance threshold, the troubleshooting module starts working. It conducts a comprehensive investigation of the space management data, including equipment distribution data, functional area division data, service personnel allocation data, and safety measures management data. By analyzing this data, the troubleshooting module attempts to find abnormal space management data that does not conform to the current business scenario. For example, it may be found that the utilization rate of certain equipment is too low in a specific business scenario, or that the allocation of service personnel in certain areas is unreasonable.

[0080] The parameter adjustment module adjusts the abnormal space management data identified by the troubleshooting module. This may involve operations such as reallocating equipment, adjusting functional area divisions, and optimizing service personnel job arrangements. The adjusted data is re-entered into the AI ​​model to evaluate the adjusted overall performance. If the output overall performance is still lower than the preset overall performance threshold, the parameter adjustment module will continue to adjust until the overall performance is no lower than the threshold. This process may require multiple iterations to ensure the effectiveness of the adjustment.

[0081] The execution module performs actual space management operations on the hotel based on the adjusted space management data. This includes rearranging equipment, adjusting the layout of functional areas, arranging service staff positions, etc. The execution module ensures that the adjusted space management plan can be effectively implemented in the actual operation of the hotel, thereby improving the overall performance of the hotel. Through the collaborative work of these four sub-modules, the decision adjustment module can continuously optimize the hotel's space management to adapt to different business scenarios, improve the hotel's operating efficiency, service quality and customer satisfaction, and ensure that the hotel always maintains a high level of operation.

[0082] Furthermore, when the comprehensive performance has not reached a preset comprehensive performance threshold, the decision adjustment module notifies the management personnel to perform manual space management on the hotel.

[0083] In this embodiment, when the artificial intelligence model and the automatic adjustment module still cannot make the hotel's comprehensive performance reach the preset threshold after multiple attempts, notifying the management staff to intervene is an effective response strategy. With their rich experience, in-depth understanding of hotel operations and flexible decision-making ability, managers can conduct more comprehensive analysis and processing of complex situations. After receiving the notification, managers can intervene in the hotel's space management from multiple angles. First, they can conduct a more in-depth review of the space management data, not only limited to the investigation of abnormal data, but also comprehensively consider the overall operation strategy, market demand changes, etc. For example, managers may analyze the current market competition situation and determine whether it is necessary to adjust the hotel's positioning and service features, thereby affecting the direction of space management. Secondly, managers can communicate and coordinate with various departments to understand the problems and difficulties in actual operations. Communicate with service personnel to understand the bottlenecks and needs they encounter in their work, so as to better adjust personnel allocation; cooperate with equipment maintenance personnel to ensure the normal operation and reasonable layout of equipment; discuss with the marketing team to adjust the setting of functional areas according to market demand.

[0084] Managers can also conduct field visits to personally experience the hotel's spatial layout and operations. They can observe guests' behaviors and needs and identify potential problem areas. For example, if they find that the passenger flow in a functional area is too large, resulting in a decline in service quality, they can consider replanning the layout of the functional area or adding service staff. Finally, after performing manual space management, managers can feed back the adjusted results to the decision adjustment module so that the module can learn and optimize its adjustment strategy. At the same time, managers can also re-evaluate and adjust the preset comprehensive performance thresholds based on this experience to make them more in line with the actual operation of the hotel.

[0085] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technical personnel in this field can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. A large-scale hotel operation and maintenance management system based on BIM and the Internet of Things, characterized by: It includes a scenario recognition module, a BIM module, a performance analysis module and a decision adjustment module, wherein the scenario recognition module, the BIM module, the performance analysis module and the decision adjustment module are communicatively connected, wherein: The scenario recognition module is used to recognize the current hotel's business scenario data; The BIM module is used to obtain the current hotel's space management data in real time through BIM; The performance analysis module is used to analyze the comprehensive performance of the hotel in the current business scenario based on the current hotel space management data, and includes the following steps: S1. Collect space management data and comprehensive performance data under different business scenarios in historical periods. The comprehensive performance data is comprehensively evaluated through the operational efficiency, service quality and customer satisfaction of large hotels; S2, extract the features of space management data, and use the comprehensive performance data as the corresponding labels to train the artificial intelligence model offline; S3: Input the current hotel's space management data into the trained artificial intelligence model and output the comprehensive performance under the current business scenario; The decision adjustment module is used to continuously adjust the hotel's space management according to the comprehensive performance under the current business scenario until a preset comprehensive performance threshold is reached.

2. A large-scale hotel operation and maintenance management system based on BIM and the Internet of Things according to claim 1, characterized in that: In step S1, the comprehensive evaluation of the operation efficiency, service quality and customer satisfaction of large hotels includes the following steps: S11. Determine the evaluation index system: clarify the evaluation indexes of operation efficiency, service quality and customer satisfaction, among which: the evaluation indexes of operation efficiency include equipment utilization rate, staff work efficiency and energy consumption rate; the evaluation indexes of service quality include the integrity of facilities and equipment, the standardization and efficiency of service processes and the degree of personalization of services; the evaluation indexes of customer satisfaction include the guests' satisfaction evaluation of the overall environment of the hotel, service attitude, catering quality and accommodation comfort; S12. Quantitative evaluation: Quantify the evaluation indicators and obtain the comprehensive evaluation results of operational efficiency, service quality and customer satisfaction; S13. Determine weights: Determine the weights of operational efficiency, service quality and customer satisfaction based on the hotel’s actual situation and development priorities; S14. Comprehensive calculation: Based on the determined weights and quantified comprehensive evaluation results, a comprehensive calculation is performed to obtain the hotel's comprehensive performance score.

3. A large-scale hotel operation and maintenance management system based on BIM and the Internet of Things according to claim 1, characterized in that: In step S2, extracting the space management data features includes the following steps: Numerical processing of spatial management data is performed to obtain multi-dimensional spatial management variable data; The principal component analysis method is used to reduce the dimension of multidimensional space management variable data to obtain a number of space management principal components, which are used as space management data features.

4. A large-scale hotel operation and maintenance management system based on BIM and the Internet of Things according to claim 3, characterized in that: The spatial management principal component is used as the spatial management data feature, specifically, the front part with a total explained variance greater than 80% is selected. n The spatial management principal components are used as the spatial management data features.

5. According to a large-scale hotel operation and maintenance management system based on BIM and the Internet of Things according to claim 1, it is characterized in that: In step S2, the artificial intelligence model is configured as a multi-layer perceptron, comprising the following steps: S21, data preparation: sort out the extracted spatial management data features and the corresponding comprehensive performance data labels, and divide the data samples into training set, validation set and test set; S22. Network structure design: determine the input layer, hidden layer and output layer neuron composition of the multi-layer perceptron; S23, Initialize weights and biases: assign initial values ​​to each connection weight and bias in the network; S24, forward propagation: Input a sample data in the training set into the multi-layer perceptron, and perform forward propagation through the calculation of each layer. In each layer, the input and weight are linearly combined, and the bias is added. Then, the output of each layer is obtained through the activation function, and the prediction result of the output layer is finally obtained after passing through the input layer and the hidden layer in turn. S25. Calculate loss: Compare the prediction result of the output layer with the actual comprehensive performance data label and calculate the value of the loss function; S26, back propagation: According to the value of the loss function, the gradient of the weights and biases of each layer is calculated through the back propagation algorithm; S27, parameter update: use the optimization algorithm to update the weights and biases in the network according to the calculated gradient; S28, model training and adjustment: repeat steps S24 to S27, perform multiple iterations of training on the entire training set, and during the training process, regularly use the validation set to evaluate the performance of the model, adjust the model's hyperparameters based on the results of the validation set, and stop training when the performance of the model on the validation set no longer improves; S29, Model evaluation: Use the test set to perform a final evaluation of the trained model.

6. A large-scale hotel operation and maintenance management system based on BIM and the Internet of Things according to claim 1, characterized in that: The decision adjustment module includes a monitoring module, a troubleshooting module, a parameter adjustment module and an execution module, wherein: The monitoring module is used to monitor and analyze whether the comprehensive performance under the current business scenario is lower than a preset comprehensive performance threshold; The troubleshooting module is used to troubleshoot the space management data when the comprehensive performance is lower than a preset comprehensive performance threshold, and determine abnormal space management data that does not conform to the current business scenario; The parameter adjustment module is used to adjust the abnormal space management data and re-input the artificial intelligence model until the output comprehensive performance is not lower than the preset comprehensive performance threshold, and the adjustment is completed; The execution module performs space management on the hotel according to the adjusted space management data.

7. A large-scale hotel operation and maintenance management system based on BIM and the Internet of Things according to claim 1, characterized in that: The decision adjustment module notifies the management personnel to perform manual space management on the hotel when the comprehensive performance has not reached a preset comprehensive performance threshold.

8. A large-scale hotel operation and maintenance management system based on BIM and the Internet of Things according to claim 1, characterized in that: The business scenario data includes customer flow, weather changes, seasonal changes and special events.

9. A large-scale hotel operation and maintenance management system based on BIM and the Internet of Things according to claim 1, characterized in that: The space management data includes equipment distribution data, functional area division data, service personnel allocation data and safety measures management data.

10. A large-scale hotel operation and maintenance management system based on BIM and the Internet of Things according to claim 1, characterized in that: The spatial management data is visualized using GIS.

Images