Intelligent hotel guest room scheduling system and hotel management method

Abstract

The invention discloses an intelligent hotel guest room scheduling system and a hotel management method, and relates to the technical field of hotel management. The system comprises a central cloud, an edge node, hotel management equipment and an intelligent terminal, the central cloud comprises a scheduling layer and a data layer; the edge node comprises a decision layer, a sample layer and an acquisition layer; the decision-making layer comprises a scheduling model and an evaluation function algorithm, and the sample layer comprises a generation module and a discrimination module; the hotel management device is located at a hotel reception desk and is used for carrying out information collection and management and control on each guest room of the hotel. The intelligent terminal is located in a hotel guest room. According to the invention, the Internet of Things for connecting the intelligent terminals in the whole hotel is constructed, the sample data set is collected to train the scheduling model, the scheduling model outputs the auxiliary decision scheduling result and feeds back the result to the hotel management equipment, and the hotel management equipment completes the scheduling and distribution work of the rooms, thereby improving the intelligent degree and check-in efficiency of the hotel, and improving the efficiency of the hotel. And the hotel management cost is reduced.

Description

Technical Field

[0001] The invention belongs to the technical field of hotel management, and in particular relates to an intelligent hotel guest room scheduling system and a hotel management method. Background Art

[0002] At present, smart tourism, also known as intelligent tourism, is developing rapidly, and travelers are increasingly demanding hotel services. With the support of new technologies such as cloud computing and the Internet of Things, through the Internet / mobile Internet, with the help of portable mobile terminal Internet devices, information on tourism resources, tourism economy, tourism activities, tourists, etc. can be actively sensed and released in a timely manner, so that people can understand this information in a timely manner, arrange and adjust work and travel plans in a timely manner, thereby achieving the effect of intelligent perception and convenient use of various tourism information.

[0003] Currently, most traditional hotels require users to check in directly at the hotel or make reservations through the website and then check in at the hotel. This has technical problems such as complicated check-in procedures, inefficient check-in services, and high human resource costs for hotel management. In addition, registered users are unable to remotely control the equipment in the guest rooms, and the level of intelligent services in hotels is relatively low. Summary of the invention

[0004] The purpose of the present invention is to provide an intelligent hotel guest room scheduling system and a hotel management method. By constructing an Internet of Things in which various intelligent terminals in the entire hotel are connected to each other, a sample data set is collected to train a scheduling model, and the scheduling model outputs an auxiliary decision-making scheduling result and feeds it back to the hotel management equipment. The hotel management equipment completes the scheduling and allocation of the room, thereby solving the problems of low intelligence level, high management cost and low efficiency of tourist check-in in existing hotels.

[0005] In order to solve the above technical problems, the present invention is achieved through the following technical solutions:

[0006] The present invention is an intelligent hotel room scheduling system, comprising a central cloud, an edge node, a hotel management device and an intelligent terminal;

[0007] The central cloud includes a scheduling layer and a data layer; the scheduling layer is a scheduling module that is responsible for data collection, model iteration and data distribution within the domain, and completes the overall scheduling of hotel guest rooms; the data layer includes an aggregation network module and a synthetic sample module; the aggregation network module is used to set up an aggregation network database, store the scheduling model parameter information uploaded by the nodes in the region, and realize model fusion through calculation; the synthetic sample module is used to set up a synthetic sample set database, store the sample data uploaded by the nodes in the region, and add data disturbance and feature noise;

[0008] The edge node includes a decision layer, a sample layer and a collection layer; the decision layer includes a scheduling model and an evaluation function algorithm, which is used to generate a hotel room scheduling decision output based on the synthetic sample; the sample layer includes a generation module and a discrimination module; the discrimination module is used to classify and fit the data sample; the generation module is used to superimpose the fitting data and the data sample into a synthetic set and send it to the data layer; the collection layer is used to obtain the hotel's historical hotel room information and occupant information;

[0009] The hotel management equipment is located at the hotel front desk and is used to collect and control information on each guest room in the hotel;

[0010] The intelligent terminal is located in the hotel guest room, and specifically includes a guest room call terminal, an intelligent door lock, a guest room monitoring device and a service terminal.

[0011] As an optimal technical solution, the hotel management device includes a hotel room display module, a storage module, a hotel room reservation module, a hotel room check-in module, and a hotel room check-out module; the hotel room display module is connected to the data acquisition module, and is used to display the basic information, photos and status of all hotel rooms; the storage module is connected to the data acquisition module, and is used to store the basic information, photos and status of the hotel rooms, and store the customer check-in information, check-out information and reservation information of the hotel rooms; the hotel room reservation module is connected to the storage module, and is used to make reservations for rooms with a vacant status according to the basic information and photos of the hotel rooms, and generate reservation information; the hotel room check-in module is respectively connected to the storage module and the hotel room reservation module, and is used to fill in the customer check-in information according to the reservation information, and update the room status to checked-in; the hotel room check-out module is respectively connected to the storage module and the hotel room reservation module, and is used to fill in the customer check-out information according to the reservation information and the current check-out time, and update the room status to vacant.

[0012] As a preferred technical solution, the sampling layer is connected to the hotel management equipment through the Internet of Things intelligent gateway; the hotel management equipment collects the status of each smart device in each guest room of the hotel, the power consumption of each device in the guest room, the temperature and humidity status in the guest room, as well as the room type, area, cost, number of occupants and address.

[0013] As a preferred technical solution, when the synthetic sample module sets up the synthetic sample set database, it constructs a sample sharing platform while providing data differential privacy protection for nodes within the domain; the sample sharing platform is used to provide private data sample resources for the scheduling model.

[0014] The present invention is a method for scheduling and managing guest rooms in an intelligent hotel, comprising the following steps:

[0015] Step S1: Based on the monitoring event category and data characteristics, a large amount of hotel room sensor information, hotel room information and hotel occupant information of edge nodes are obtained to produce a real sample set;

[0016] Step S2: preprocessing the sample set to produce a sample set;

[0017] Step S3: Upload the sample set to the data layer to synthesize the sample set data;

[0018] Step S4: The aggregation network module uses the sample data set to train the scheduling model and uploads it to the decision layer after the training is completed;

[0019] Step S5: The decision layer adjusts the scheduling model parameters to form an aggregation network, and completes the machine learning optimization iteration after analysis, evaluation and weighted fusion;

[0020] Step S6: The hotel management device obtains the user's reservation information and inputs it into the scheduling model optimized by the decision-making layer;

[0021] Step S7: The scheduling model outputs the auxiliary decision scheduling result and feeds the result back to the hotel management device, which completes the scheduling allocation of the rooms.

[0022] As a preferred technical solution, in step S1, the specific process of obtaining edge node information is as follows:

[0023] Step S11: Create access services for various hotel devices and complete the development of encoding and decoding plug-ins for device models;

[0024] Step S12: Build a smart hotel model on the Internet of Things platform to connect the Internet of Things platform with various hotel devices; the devices use the MQTT protocol to report data, and you need to install an MQTT client such as MQTT; configure authentication parameters in the MQTT client, including Broker Address, Broker Port southbound MQTT device access port, such as 7883, ClientId, Username and Password.

[0025] Step S13: Add the edge device that needs to report data on the edge node and configure relevant parameters, such as the device ID, name, key, etc.;

[0026] Step S14: When collecting data, the edge node collects various types of device data and performs preliminary processing, including filtering, aggregation, etc.;

[0027] Step S15: The processed data is transmitted from the edge node to the central cloud for analysis and storage to reduce the amount of data transmission and improve the response speed; when storing, the processed data is transmitted from the edge node to the central cloud or cloud platform for further analysis and storage to ensure the security and reliability of data transmission, and adopt security measures such as encrypted communication protocols.

[0028] As a preferred technical solution, in step S2, data collection and preprocessing include:

[0029] Data collection: Collect the hotel’s historical room booking data, including check-in time, check-out time, room type, price, customer preferences and other information. This data can be obtained through channels such as hotel management system (HMS) and online booking platforms.

[0030] Data cleaning: Clean the collected data to remove outliers, missing values ​​and duplicate records to ensure the accuracy and completeness of the data.

[0031] Feature Engineering: Extract useful features from raw data, such as room occupancy, average length of stay, customer satisfaction, etc., so that the model can better understand and learn from the data.

[0032] As a preferred technical solution, in step S3, the sample set data layer constructs a virtual sample set based on the sample set using a low-dimensional model generator, and improves learning ability under the reverse training of the discriminator to reduce the loss difference between the sample set and the virtual sample set. The edge node performs a union process on the locally passed sample set and the virtual sample set, and uploads the synthesized sample set data to the data layer, which solves the problem of inaccurate model training due to small data volume, eliminates sample sets with potential differences, enriches and completes synthetic samples, and supports the next step of machine learning.

[0033] As a preferred technical solution, in step S4, the scheduling model training process is as follows:

[0034] Step S41: extracting useful features from the sample set data;

[0035] Step S42: According to the nature of room scheduling and data characteristics, the deep learning model is selected as a convolutional neural network model, and a regularized linear unit function is used as an activation function;

[0036] Step S43: setting training parameters such as learning rate, batch size, iteration parameters, etc., and performing scheduling model training;

[0037] Step S44: Input the features of the sample set into the model, obtain the prediction result through forward propagation, and use the loss function to calculate the gap between the prediction result and the real sample;

[0038] Step S45: According to the value of the loss function, the gradient is calculated by the back propagation algorithm, and the model parameters are updated to minimize the loss function;

[0039] Step S46: Evaluate the performance of the model on the validation set, and adjust the model structure or training parameters according to the evaluation results to complete the scheduling model training.

[0040] After the scheduling model training is completed, the trained model needs to be deployed to the production environment for real-time room scheduling and reservation management, and the performance and effect of the model should be continuously monitored. The model should be regularly checked for overfitting or underfitting, and the model should be updated or retrained as needed.

[0041] As a preferred technical solution, in step S44, the loss function is used to represent the gap between the model prediction result and the real sample, and specifically adopts the cross entropy loss function, and the specific formula is as follows:

[0042]

[0043] In the formula, L is the loss value, which is used to measure the gap between the model prediction result and the true label. The smaller the loss value, the better the performance of the model, n is the number of samples, y is i Indicates the value of the i-th category of real data, ranging from 0 to 1, p i Represents the probability of the i-th category of the model. This probability is usually obtained by passing the model's output through a softmax function or other normalization process to ensure that the sum of the probabilities of all categories is 1;

[0044] The cross entropy loss function evaluates the performance of the model by comparing the difference between the probability distribution predicted by the model and the true label, and guides the update of the model parameters. During the training process, the optimizer adjusts the model parameters according to the gradient calculated by the cross entropy loss function to minimize the loss value, thereby improving the prediction accuracy of the model.

[0045] As a preferred technical solution, in step S7, the scheduling model predicts the time distribution of users arriving at the hotel, and its probability density function is as follows:

[0046]

[0047] In the formula, x is the time when the user arrives at the hotel, μ is the mean of the user's lead time, and σ 2 is the variance of the user's lead time;

[0048] The predicted number of users arriving at the hotel during the specified time period is:

[0049]

[0050] In the formula, is the number of people who arrive at the i-th hotel during the time period from t1 to t2, N is the total number of people who book the hotel, t1 is the start time of the prediction time period, and t2 is the end time of the prediction time period.

[0051] The present invention has the following beneficial effects:

[0052] The present invention constructs an Internet of Things in which various intelligent terminals in the entire hotel are connected to each other, collects sample data sets to train a scheduling model, and the scheduling model outputs auxiliary decision-making scheduling results and feeds them back to the hotel management equipment. The hotel management equipment completes the scheduling and allocation of rooms, thereby improving the hotel's intelligence level and occupancy efficiency, while reducing hotel management costs.

[0053] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0055] Figure 1 A flow chart of a smart hotel guest room scheduling management method of the present invention;

[0056] Figure 2 The present invention is a schematic diagram of the structure of an intelligent hotel guest room scheduling system. DETAILED DESCRIPTION

[0057] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0058] In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0059] In order to make the purpose, technical solutions and advantages of this application more clear, the following Figure 1-2 It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0060] Embodiment 1

[0061] See also Figure 1 As shown, the present invention is a smart hotel room scheduling management method, comprising the following steps:

[0062] Step S1: Based on the monitoring event category and data characteristics, a large amount of hotel room sensor information, hotel room information and hotel occupant information of edge nodes are obtained to produce a real sample set;

[0063] Step S2: preprocessing the sample set to produce a sample set;

[0064] Step S3: Upload the sample set to the data layer to synthesize the sample set data;

[0065] Step S4: The aggregation network module uses the sample data set to train the scheduling model and uploads it to the decision layer after the training is completed;

[0066] Step S5: The decision layer adjusts the scheduling model parameters to form an aggregation network, and completes the machine learning optimization iteration after analysis, evaluation and weighted fusion;

[0067] Step S6: The hotel management device obtains the user's reservation information and inputs it into the scheduling model optimized by the decision-making layer;

[0068] Step S7: The scheduling model outputs the auxiliary decision scheduling result and feeds the result back to the hotel management device, which completes the scheduling allocation of the rooms.

[0069] In step S1, the specific process of obtaining edge node information is as follows:

[0070] Step S11: Create access services for various hotel devices and complete the development of encoding and decoding plug-ins for device models;

[0071] Step S12: Build a smart hotel model on the Internet of Things platform to connect the Internet of Things platform with various hotel devices; the devices use the MQTT protocol to report data, and you need to install an MQTT client such as MQTT; configure authentication parameters in the MQTT client, including Broker Address, Broker Port southbound MQTT device access port, such as 7883, ClientId, Username and Password.

[0072] Step S13: Add the edge device that needs to report data on the edge node and configure relevant parameters, such as the device ID, name, key, etc.;

[0073] Step S14: When collecting data, the edge node collects various types of device data and performs preliminary processing, including filtering, aggregation, etc.;

[0074] Step S15: The processed data is transmitted from the edge node to the central cloud for analysis and storage to reduce the amount of data transmission and improve the response speed; when storing, the processed data is transmitted from the edge node to the central cloud or cloud platform for further analysis and storage to ensure the security and reliability of data transmission, and adopt security measures such as encrypted communication protocols.

[0075] In step S2, data collection and preprocessing include:

[0076] Data collection: Collect the hotel’s historical room booking data, including check-in time, check-out time, room type, price, customer preferences and other information. This data can be obtained through channels such as hotel management system (HMS) and online booking platforms.

[0077] Data cleaning: Clean the collected data to remove outliers, missing values ​​and duplicate records to ensure the accuracy and completeness of the data.

[0078] Feature Engineering: Extract useful features from raw data, such as room occupancy, average length of stay, customer satisfaction, etc., so that the model can better understand and learn from the data.

[0079] In step S3, the sample set data layer constructs a virtual sample set based on the sample set using a low-dimensional model generator, and improves learning ability under the reverse training of the discriminator to reduce the loss difference between the sample set and the virtual sample set. The edge node combines the local sample set and the virtual sample set and uploads them to the data layer to synthesize the sample set data, which solves the problem of inaccurate model training due to small data volume, eliminates sample sets with potential differences, enriches and completes synthetic samples, and supports the next step of machine learning.

[0080] In step S4, the scheduling model training process is as follows:

[0081] Step S41: extracting useful features from the sample set data;

[0082] Step S42: According to the nature of room scheduling and data characteristics, the deep learning model is selected as a convolutional neural network model, and a regularized linear unit function is used as an activation function;

[0083] Step S43: setting training parameters such as learning rate, batch size, iteration parameters, etc., and performing scheduling model training;

[0084] Step S44: Input the features of the sample set into the model, obtain the prediction result through forward propagation, and use the loss function to calculate the gap between the prediction result and the real sample;

[0085] Step S45: According to the value of the loss function, the gradient is calculated by the back propagation algorithm, and the model parameters are updated to minimize the loss function;

[0086] Step S46: Evaluate the performance of the model on the validation set, and adjust the model structure or training parameters according to the evaluation results to complete the scheduling model training.

[0087] After the scheduling model training is completed, the trained model needs to be deployed to the production environment for real-time room scheduling and reservation management, and the performance and effect of the model should be continuously monitored. The model should be regularly checked for overfitting or underfitting, and the model should be updated or retrained as needed.

[0088] Step S44: The loss function is used to represent the gap between the model prediction result and the real sample. Specifically, the cross entropy loss function is used. The specific formula is as follows:

[0089]

[0090] In the formula, L is the loss value, which is used to measure the gap between the model prediction result and the true label. The smaller the loss value, the better the performance of the model, n is the number of samples, y is i Indicates the value of the i-th category of real data, ranging from 0 to 1, p i Represents the probability of the i-th category of the model. This probability is usually obtained by passing the model's output through a softmax function or other normalization process to ensure that the sum of the probabilities of all categories is 1;

[0091] The cross entropy loss function evaluates the performance of the model by comparing the difference between the probability distribution predicted by the model and the true label, and guides the update of the model parameters. During the training process, the optimizer adjusts the model parameters according to the gradient calculated by the cross entropy loss function to minimize the loss value, thereby improving the prediction accuracy of the model.

[0092] In step S7, the scheduling model predicts the time distribution of users arriving at the hotel, and its probability density function is as follows:

[0093]

[0094] In the formula, x is the time when the user arrives at the hotel, μ is the mean of the user's lead time, and σ 2 is the variance of the user's lead time;

[0095] The predicted number of users arriving at the hotel during the specified time period is:

[0096]

[0097] In the formula, is the number of people who arrive at the i-th hotel during the time period from t1 to t2, N is the total number of people who book the hotel, t1 is the start time of the prediction time period, and t2 is the end time of the prediction time period.

[0098] Embodiment 2

[0099] See also Figure 2 As shown, the present invention is an intelligent hotel room scheduling system, which can be used to execute the method content of embodiment 1 of the present invention, including: a central cloud, an edge node, a hotel management device and an intelligent terminal;

[0100] The central cloud includes a scheduling layer and a data layer; the scheduling layer is a scheduling module that is responsible for data collection, model iteration and data distribution within the domain, and completes the overall scheduling of hotel guest rooms; the data layer includes an aggregation network module and a synthetic sample module; the aggregation network module is used to set up an aggregation network database, store the scheduling model parameter information uploaded by nodes in the region, and realize model fusion through calculation; the synthetic sample module is used to set up a synthetic sample set database, store sample data uploaded by nodes in the region, and add data disturbance and feature noise;

[0101] The edge node includes a decision layer, a sample layer, and a collection layer; the decision layer includes a scheduling model and an evaluation function algorithm, which are used to generate hotel room scheduling decision outputs based on synthetic samples; the sample layer includes a generation module and a discrimination module; the discrimination module is used to classify and fit data samples; the generation module is used to superimpose the fitting data and data samples into a synthetic set and send it to the data layer; the collection layer is used to obtain the hotel's historical hotel room information and occupant information;

[0102] The hotel management equipment is located at the hotel front desk and is used to collect information and control each guest room in the hotel; the smart terminal is located in the hotel guest room, specifically including the guest room call terminal, smart door lock, guest room monitoring equipment and service terminal. Each guest room is equipped with a guest room call terminal, smart door lock and guest room monitoring equipment. The guest room call terminal is divided into fixed and mobile types. The fixed guest room call terminal is installed on the wall of the guest room, and the mobile guest room call terminal can be an APP or a web page. The customer can install or open the mobile guest room call terminal. The guest room call terminal can receive calls from customers, and after determining the specific call content, it sends the corresponding call information to the mobile service terminal to remind the service staff to go and handle it.

[0103] The hotel management equipment includes a hotel room display module, a storage module, a hotel room reservation module, a hotel room check-in module, and a hotel room check-out module; the hotel room display module is connected to the data acquisition module, and is used to display the basic information, photos and status of all hotel rooms; the storage module is connected to the data acquisition module, and is used to store the basic information, photos and status of the hotel rooms, and store the customer check-in information, check-out information and reservation information of the hotel rooms; the hotel room reservation module is connected to the storage module, and is used to reserve rooms with a vacant status based on the basic information and photos of the hotel rooms, and generate reservation information; the hotel room check-in module is respectively connected to the storage module and the hotel room reservation module, and is used to fill in the customer check-in information based on the reservation information, and update the room status to checked-in; the hotel room check-out module is respectively connected to the storage module and the hotel room reservation module, and is used to fill in the customer check-out information based on the reservation information and the current time of check-out, and update the room status to vacant.

[0104] The electronic device controller inside each guest room of the hotel may include at least one of the intelligent terminal control units such as the guest room lighting control unit, the guest room air conditioning control unit, the guest room curtain control unit, the guest room smart TV control unit, the guest room hot water supply control unit, the guest room house number display unit, and the guest room door lock control unit. Each intelligent terminal control unit can be set on its corresponding electronic device or in a line bin inside the guest room wall.

[0105] The smart terminal control unit in each guest room can control its corresponding smart terminal. For example, the guest room lighting control unit is connected to the guest room lighting to control the status of the guest room lighting; the guest room air conditioning control unit is connected to the guest room air conditioning to control the status of the guest room air conditioning; the guest room curtain control unit is connected to the guest room curtain to control the status of the guest room curtain; the guest room smart TV control unit is connected to the guest room smart TV to control the status of the guest room smart TV.

[0106] The sampling layer is connected to the hotel management equipment through the IoT smart gateway; the hotel management equipment collects the status of each smart device in each guest room of the hotel, the power consumption of each device in the guest room, the temperature and humidity status in the guest room, as well as the room type, area, fee, number of occupants and address.

[0107] When the synthetic sample module sets up the synthetic sample collection database, it constructs a sample sharing platform while providing data differential privacy protection for nodes in the domain; the sample sharing platform is used to provide private data sample resources for the scheduling model.

[0108] It is worth noting that in the above system embodiment, the various units included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0109] In addition, those skilled in the art can understand that all or part of the steps in the above-mentioned embodiments can be completed by instructing related hardware through a program, and the corresponding program can be stored in a computer-readable storage medium.

[0110] The preferred embodiments of the present invention disclosed above are only used to help illustrate the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. An intelligent hotel room scheduling system, including a central cloud, edge nodes, hotel management equipment and intelligent terminals, characterized in that: The central cloud includes a scheduling layer and a data layer; the scheduling layer is a scheduling module that is responsible for data collection, model iteration and data distribution within the domain, and completes the overall scheduling of hotel guest rooms; the data layer includes an aggregation network module and a synthetic sample module; the aggregation network module is used to set up an aggregation network database, store the scheduling model parameter information uploaded by the nodes in the region, and realize model fusion through calculation; the synthetic sample module is used to set up a synthetic sample set database, store the sample data uploaded by the nodes in the region, and add data disturbance and feature noise; The edge node includes a decision layer, a sample layer and a collection layer; the decision layer includes a scheduling model and an evaluation function algorithm, which is used to generate a hotel room scheduling decision output based on the synthetic sample; the sample layer includes a generation module and a discrimination module; the discrimination module is used to classify and fit the data sample; the generation module is used to superimpose the fitting data and the data sample into a synthetic set and send it to the data layer; the collection layer is used to obtain the hotel's historical hotel room information and occupant information; The hotel management equipment is located at the hotel front desk and is used to collect and control information on each guest room in the hotel; The intelligent terminal is located in the hotel guest room, and specifically includes a guest room call terminal, an intelligent door lock, a guest room monitoring device and a service terminal.

2. An intelligent hotel room scheduling system according to claim 1, characterized in that: The hotel management device includes a hotel room display module, a storage module, a hotel room reservation module, a hotel room check-in module, and a hotel room check-out module; the hotel room display module is connected to the data acquisition module, and is used to display the basic information, photos and status of all hotel rooms; the storage module is connected to the data acquisition module, and is used to store the basic information, photos and status of the hotel rooms, and store the customer check-in information, check-out information and reservation information of the hotel rooms; the hotel room reservation module is connected to the storage module, and is used to make reservations for rooms with a vacant status according to the basic information and photos of the hotel rooms, and generate reservation information; the hotel room check-in module is respectively connected to the storage module and the hotel room reservation module, and is used to fill in the customer check-in information according to the reservation information, and update the room status to checked-in; the hotel room check-out module is respectively connected to the storage module and the hotel room reservation module, and is used to fill in the customer check-out information according to the reservation information and the current check-out time, and update the room status to vacant.

3. An intelligent hotel room scheduling system according to claim 1, characterized in that: The sampling layer is connected to the hotel management equipment through the Internet of Things intelligent gateway; the hotel management equipment collects the status of each intelligent device in each guest room of the hotel, the power consumption of each device in the guest room, the temperature and humidity status in the guest room, as well as the room type, area, cost, number of occupants and address.

4. An intelligent hotel room scheduling system according to claim 1, characterized in that: When the synthetic sample module sets up the synthetic sample set database, it constructs a sample sharing platform while providing data differential privacy protection for nodes within the domain; the sample sharing platform is used to provide private data sample resources for the scheduling model.

5. A smart hotel room scheduling management method, characterized in that: The steps include: Step S1: Based on the monitoring event category and data characteristics, a large amount of hotel room sensor information, hotel room information and hotel occupant information of edge nodes are obtained to produce a real sample set; Step S2: preprocessing the sample set to produce a sample set; Step S3: Upload the sample set to the data layer to synthesize the sample set data; Step S4: The aggregation network module uses the sample data set to train the scheduling model and uploads it to the decision layer after the training is completed; Step S5: The decision layer adjusts the scheduling model parameters to form an aggregation network, and completes the machine learning optimization iteration after analysis, evaluation and weighted fusion; Step S6: The hotel management device obtains the user's reservation information and inputs it into the scheduling model optimized by the decision-making layer; Step S7: The scheduling model outputs the auxiliary decision scheduling result and feeds the result back to the hotel management device, which completes the scheduling allocation of the rooms.

6. A smart hotel room scheduling management method according to claim 5, characterized in that: In step S1, the specific process of obtaining edge node information is as follows: Step S11: Create access services for various hotel devices and complete the development of encoding and decoding plug-ins for device models; Step S12: constructing a smart hotel model on the Internet of Things platform to connect the Internet of Things platform with various hotel devices; Step S13: Add edge devices that need to report data to the edge node and configure relevant parameters; Step S14: When collecting data, the edge node collects data from various devices and performs preliminary processing; Step S15: The processed data is transmitted from the edge node to the central cloud for analysis and storage.

7. A smart hotel room scheduling management method according to claim 5, characterized in that: In step S3, the sample set data layer constructs a virtual sample set based on the sample set using a low-dimensional model generator, and improves the learning ability under the reverse training of the discriminator. The edge node performs a union process on the local sample set and the virtual sample set, and uploads them to the data layer to synthesize the sample set data.

8. A smart hotel room scheduling management method according to claim 5, characterized in that: In step S4, the scheduling model training process is as follows: Step S41: extracting useful features from the sample set data; Step S42: According to the nature of room scheduling and data characteristics, the deep learning model is selected as a convolutional neural network model, and a regularized linear unit function is used as an activation function; Step S43: setting the learning rate, batch size, and iteration parameters to perform scheduling model training; Step S44: Input the features of the sample set into the model, obtain the prediction result through forward propagation, and use the loss function to calculate the gap between the prediction result and the real sample; Step S45: According to the value of the loss function, the gradient is calculated by the back propagation algorithm, and the model parameters are updated to minimize the loss function; Step S46: Evaluate the performance of the model on the validation set, and adjust the model structure or training parameters according to the evaluation results to complete the scheduling model training.

9. A smart hotel room scheduling management method according to claim 8, characterized in that: In step S44, the loss function is used to represent the gap between the model prediction result and the real sample, and specifically adopts the cross entropy loss function, and the specific formula is as follows: In the formula, L is the loss value, n is the number of samples, and y i represents the value of the i-th category of real data, p i Represents the probability of the i-th category of the model.

10. A smart hotel room scheduling management method according to claim 5, characterized in that: In step S7, the scheduling model predicts the time distribution of users arriving at the hotel, and its probability density function is as follows: In the formula, x is the time when the user arrives at the hotel, μ is the mean of the user's lead time, and σ 2 is the variance of the user's lead time; The predicted number of users arriving at the hotel during the specified time period is: In the formula, is the number of people who arrive at the i-th hotel during the time period from t1 to t2, N is the total number of people who book the hotel, t1 is the start time of the prediction time period, and t2 is the end time of the prediction time period.

Images