Hotel data monitoring method, device, equipment and medium

Abstract

The invention discloses a hotel data monitoring method and device, equipment and a medium, and the method comprises the steps: obtaining function data corresponding to a plurality of hotel function systems based on a data summarization platform, and storing the function data of the plurality of hotel function systems; the state monitoring system determines whether a target system with an abnormal state exists in the plurality of hotel function systems based on various function data stored in the data summarization platform; and under the condition that a target system with an abnormal state exists in the plurality of hotel function systems, the state monitoring system obtains to-be-analyzed data related to the target system from the data summarization platform, determines an abnormal source causing the abnormal state of the target system based on the to-be-analyzed data, and outputs abnormal alarm information according to the abnormal source. According to the invention, state monitoring of all management and control systems of the hotel can be coped with, the state monitoring system identifies system abnormity in real time, and compared with traditional threshold warning, early warning is carried out in advance, the average response time can be shortened to the second level, and the response speed is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, and in particular to a hotel data monitoring method, device, equipment and medium. BACKGROUND

[0002] With the increasing demand of people for travel and accommodation, the types and quantities of hotel management and control systems are also increasing, and the hotel management and control systems are becoming more and more complex, with numerous system components and intensive dependency relationships. When the system is abnormal, the traditional operation and maintenance method relies on manual log and index checking, which is low in efficiency, slow in response, and prone to omissions and false reports. Therefore, how to monitor the abnormality of the hotel management and control systems with various types and quantities is still a problem to be solved at present. SUMMARY

[0003] The embodiments of the present application provide a hotel data monitoring method, device, equipment and medium, which solve the technical problems in the prior art that hotel system abnormalities rely on manual log and index checking, which is low in efficiency, slow in response, and prone to omissions and false reports, and achieve the technical effect of monitoring the abnormality of the hotel management and control systems with various types and quantities.

[0004] In a first aspect, the present application provides a hotel data monitoring method, the hotel comprising a state monitoring system, a data aggregation platform and a plurality of hotel function systems, the state monitoring system and the data aggregation platform performing information interaction, and the data aggregation platform and the plurality of hotel function systems performing information interaction; the method comprises: The data aggregation platform obtains function data corresponding to each of the plurality of hotel function systems, and stores the function data of the plurality of hotel function systems; The state monitoring system determines whether there is a target system with a state abnormality in the plurality of hotel function systems based on various function data stored by the data aggregation platform; In the case where there is a target system with a state abnormality in the plurality of hotel function systems, the state monitoring system obtains, from the data aggregation platform, to-be-analyzed data related to the target system, determines an abnormal source causing the state abnormality of the target system based on the to-be-analyzed data, and outputs an abnormal alarm information according to the abnormal source.

[0005] Further, the data aggregation platform obtains function data corresponding to each of the plurality of hotel function systems, comprising: According to a preset requirement level of the function data of each hotel function system, a preset mode of the data aggregation platform obtaining function data from each hotel function system is determined; the preset requirement level is determined according to the data change frequency, real-time requirement and data volume; If the preset demand level of the hotel function system is a high level, the preset mode corresponding to the hotel function system is that the hotel function system actively sends function data to the state monitoring system according to a first preset frequency; if the preset demand level of the hotel function system is a low level, the state monitoring system actively collects function data from the hotel function system according to a second preset frequency.

[0006] Further, the data aggregation platform acquires function data corresponding to each of the plurality of hotel function systems, and stores the function data of the plurality of hotel function systems, including: The data aggregation platform acquires original function data corresponding to each of the plurality of hotel function systems, and converts the original function data of each of the hotel function systems into benchmark function data meeting a preset benchmark data format requirement; The benchmark function data of each of the hotel function systems is stored.

[0007] Further, the state monitoring system includes a plurality of pre-trained machine learning models, and the machine learning models correspond one-to-one to the hotel function systems; The state monitoring system determines whether there is a target system with a state anomaly in the plurality of hotel function systems based on various function data stored by the data aggregation platform, including: For each of the hotel function systems, the state monitoring system inputs function data of the hotel function system in the data aggregation platform within a target historical period into the machine learning model of the hotel function system, so that the machine learning model outputs a monitoring result representing whether the hotel function system is in a state anomaly; wherein the machine learning model is trained based on function data in a state normal period before the target historical period; According to the monitoring result of each of the hotel function systems, it is determined whether there is a target system with a state anomaly in the plurality of hotel function systems.

[0008] Further, at least two hotel function systems in which function data exist mutual influence are recorded as an associated function system group; The state monitoring system determines whether there is a target system with a state anomaly in the plurality of hotel function systems based on various function data stored by the data aggregation platform, including: For each of the hotel function systems, the state monitoring system obtains an independent monitoring result representing whether the hotel function system is in a state anomaly based on function data of the hotel function system in the data aggregation platform; For each of the associated function system groups, the state monitoring system obtains an associated monitoring result representing whether the associated function system group is in an abnormal state based on the function data of each of the hotel function systems in the associated function system group in the data aggregation platform; According to the independent monitoring result and the associated monitoring result, a target system in which a state is abnormal is determined from among the plurality of hotel function systems; the target system includes the hotel function system and / or the associated function system group.

[0009] Further, the to-be-analyzed data includes function data of the target system at different time periods, function data of other systems associated with the target system, and other data associated with the target system; The determination of the abnormal source causing the state of the target system to be abnormal based on the to-be-analyzed data includes: The to-be-analyzed data is input into a pre-trained root cause probability calculation model, and a root cause probability result is output, wherein the root cause probability result includes at least one abnormal source causing the state of the target system to be abnormal and an actual probability of each abnormal source causing the state of the target system to be abnormal.

[0010] Further, the output of the abnormal alarm information according to the abnormal source includes: An abnormal alarm level is determined according to a device type corresponding to the abnormal source and an influence range; The abnormal alarm information is output in a preset alarm mode matched with the abnormal alarm level.

[0011] In a second aspect, the present application provides a hotel data monitoring device, the hotel including a state monitoring system, a data aggregation platform, and a plurality of hotel function systems, the state monitoring system and the data aggregation platform performing information interaction, and the data aggregation platform and the plurality of hotel function systems performing information interaction; the device includes: A data acquisition and storage module is configured to acquire function data corresponding to each of the plurality of hotel function systems by the data aggregation platform, and store the function data of the plurality of hotel function systems; A target system determination module is configured to determine whether a target system in which a state is abnormal exists from among the plurality of hotel function systems based on various function data stored by the data aggregation platform by the state monitoring system; An abnormal response module is configured to, in a case where the target system in which the state is abnormal exists from among the plurality of hotel function systems, obtain to-be-analyzed data related to the target system from the data aggregation platform by the state monitoring system, determine an abnormal source causing the state of the target system to be abnormal based on the to-be-analyzed data, and output abnormal alarm information according to the abnormal source.

[0012] In a third aspect, the present application provides an electronic device, comprising: a processor; a memory for storing instructions executable by the processor; wherein the processor is configured to execute to implement a hotel data monitoring method as provided in the first aspect.

[0013] In a fourth aspect, the present application provides a non-transitory computer-readable storage medium, when instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to implement a hotel data monitoring method as provided in the first aspect.

[0014] The one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: In the embodiments of the present application, the data aggregation platform is used to obtain the respective functional data of the plurality of hotel functional systems, and the functional data of the plurality of hotel functional systems is stored; the state monitoring system determines whether there is a target system with state anomaly in the plurality of hotel functional systems based on the various functional data stored by the data aggregation platform; in the case where there is a target system with state anomaly in the plurality of hotel functional systems, the state monitoring system obtains the to-be-analyzed data related to the target system from the data aggregation platform, determines the abnormal source causing the state anomaly of the target system based on the to-be-analyzed data, and outputs abnormal alarm information according to the abnormal source. It can be seen that the embodiments of the present application can cope with the state monitoring of all management and control systems of the hotel, the state monitoring system identifies system anomalies in real time, and compared with the traditional threshold alarm, the average response time is shortened to seconds, and the response speed is improved. Through intelligent analysis of the model, the abnormality recognition is more accurate, and the false alarm rate is significantly reduced. With the help of the call chain and the dependency topology, the root cause positioning accuracy can reach the module / interface level, and the fault handling efficiency is improved. The closed-loop process of automatic identification, positioning and notification can significantly reduce the manual participation and improve the automation degree of operation and maintenance. By discovering potential problems in advance and responding in time, the overall system failure risk is reduced, and the system stability is enhanced. BRIEF DESCRIPTION OF DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0016] Figure 1 The architecture schematic diagram of the hotel system provided in the embodiments of the present application is shown in the following figure: Figure 2A flowchart of a hotel data monitoring method provided by an embodiment of the present application is shown in the figure; Figure 3 A structural diagram of a hotel data monitoring device provided by an embodiment of the present application is shown in the figure; Figure 4 A structural diagram of an electronic device provided by an embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0017] The technical problem of the present application is solved by providing a hotel data monitoring method, which solves the technical problem of the prior art that hotel system is abnormally dependent on manual log and index checking, low efficiency, slow response, and easy to miss and misreport.

[0018] The technical solution of the present application is to solve the above technical problem, and the general idea is as follows: The present application embodiment acquires the function data corresponding to each of the plurality of hotel function systems based on the data aggregation platform, and stores the function data of the plurality of hotel function systems; the state monitoring system determines whether there is a target system with state anomaly in the plurality of hotel function systems based on the various function data stored by the data aggregation platform; in the case that there is a target system with state anomaly in the plurality of hotel function systems, the state monitoring system obtains the to-be-analyzed data related to the target system from the data aggregation platform, determines the abnormal source causing the state anomaly of the target system based on the to-be-analyzed data, and outputs an abnormal alarm information according to the abnormal source. It can be seen that the present application embodiment can cope with the state monitoring of all management and control systems of the hotel, and the state monitoring system can identify system anomalies in real time, which can shorten the average response time to seconds and improve the response speed compared with the traditional threshold alarm. Through intelligent analysis of the model, the abnormality recognition is more accurate, and the false alarm rate is significantly reduced. With the help of call chain and dependency topology, the root cause positioning accuracy can reach the module / interface level, and the fault handling efficiency is improved. The closed-loop process of automatic identification, positioning and notification can significantly reduce the manual participation and improve the automation degree of operation and maintenance. By discovering potential problems in advance and responding in time, the overall system failure risk is reduced, and the system stability is enhanced.

[0019] In order to better understand the above technical solution, the above technical solution will be described in detail in combination with the drawings of the specification and the specific embodiments.

[0020] First of all, the term "and / or" appearing in this paper is only to describe the association relationship of the associated objects, which means that there are three kinds of relationships, for example, A and / or B, which means that there are three kinds of situations, such as A alone, A and B together, and B alone. In addition, the character " / " in this paper generally represents that the front and rear associated objects are in an "or" relationship.

[0021] The hotel provided by the embodiments of the present application specifically refers to a hotel comprising a state monitoring system, a data aggregation platform and a plurality of hotel function systems. As shown in Figure 1 The hotel function systems, the data aggregation platform and the state monitoring system are connected in sequence, wherein each hotel function system is connected with the data aggregation platform. Figure 1 In the figure, only the hotel function system A, the hotel function system B and the hotel function system C are shown.

[0022] The plurality of hotel function systems specifically can include but are not limited to the following systems, such as a hotel management system (PMS, Property Management System), a central reservation system (CRS, Central Reservation System), a customer relationship management system (CRM, Customer Relationship Management), a revenue management system (RMS, Revenue Management System), a point of sale system (POS, Point of Sale), a housekeeping management system (HKMS, Housekeeping Management System), a computerized maintenance management system (CMMS, Computerized Maintenance Management), a building automation system (BAS, Building Automation System), a room control unit (RCU, Room Control Unit), an electronic lock system (ELS, Electronic Lock System), a human resource system, a financial system, a security system, etc.

[0023] The functions of the hotel management system include front desk reception, room allocation, reservation management, account processing, night audit, etc. The hotel management system is the central system of the hotel and is deeply integrated with other systems. The functions of the central reservation system include managing reservations of direct channels (official website, telephone, etc.) and connecting with reservation websites. The central reservation system synchronizes room status and prices with the PMS in real time to avoid overselling. The functions of the customer relationship management system include member management, customer data analysis and marketing activities (such as loyalty programs). The customer relationship system shares customer historical stay data with the PMS to support personalized services. The functions of the revenue management system include dynamic pricing, demand forecasting and room allocation optimization. The revenue management system obtains historical data from the PMS and pushes price strategies to the CRS.

[0024] The functions of the point-of-sale system include consumption billing of restaurants, bars, spas, etc. The point-of-sale system synchronizes bills with the PMS to realize unified settlement of room accounts. The functions of the room management system include cleaning task allocation, room status updating (such as "cleaned" and "under repair"). The room management system feeds back room status to the PMS in real time to ensure that the front desk accurately allocates rooms. The functions of the engineering operation system include equipment maintenance and work order distribution (such as air conditioner repair). The engineering operation system receives repair requests from the PMS or the HKMS and feeds back repair progress.

[0025] The functions of the building automation system include controlling public area equipment such as lights, air conditioners, and elevators. The building automation system links with the energy management system to save energy. The functions of the room control system include automation of temperature control, lights, and curtains in the room (such as power-off when unattended). The room control system synchronizes check-in status with the PMS (such as turning on the air conditioner in advance when the guest is about to arrive). The functions of the door lock system include card making and permission management (such as card swiping only during the check-in period). The door lock system is triggered by the PMS to make cards and interfaces with the public security system for identity information.

[0026] The functions of the human resource system include scheduling, attendance, and salary. The human resource system shares staff data with the PMS to arrange front desk shifts. The functions of the financial system include general ledger and cost accounting. The financial system receives daily revenue data from the PMS and synchronizes POS costs. The functions of the security system include monitoring, fire alarm, and access control. The security system links with the BAS (such as closing the elevator in case of fire).

[0027] As can be seen, the hotel function systems are diverse. In the related technology, each hotel function system is independent of each other, and each hotel function system sets a corresponding alarm threshold to realize data monitoring and management. However, for each hotel function system, it mainly relies on manual log and index checking, which is low in efficiency, slow in response, and prone to missed reports and false reports. In the related technology, it is also impossible to associate and monitor each hotel function system, resulting in chaotic and complex hotel system management, which affects the accuracy of the overall data state evaluation of the hotel.

[0028] To solve the above problems, the embodiments of the present application provide a hotel data monitoring method, the hotel includes a state monitoring system, a data aggregation platform, and a plurality of hotel function systems, the state monitoring system and the data aggregation platform interact information, the data aggregation platform and a plurality of the hotel function systems interact information; the method comprises steps S21-S23, please refer to Figure 2 .

[0029] Step S21, the data aggregation platform acquires the function data corresponding to each of the plurality of hotel function systems, and stores the function data of the plurality of hotel function systems; Step S22, the state monitoring system determines whether there is a target system with state abnormality in the plurality of hotel function systems based on the various function data stored by the data aggregation platform; Step S23, in the case where there is a target system with state abnormality in the plurality of hotel function systems, the state monitoring system obtains the to-be-analyzed data related to the target system from the data aggregation platform, determines the abnormal source causing the state abnormality of the target system based on the to-be-analyzed data, and outputs abnormal alarm information according to the abnormal source.

[0030] Regarding step S21, the data aggregation platform acquires the function data corresponding to each of the plurality of hotel function systems, and stores the function data of the plurality of hotel function systems.

[0031] The data sending and receiving relationship between the hotel function system and the data aggregation platform is that each hotel function system is a data sender, and the data aggregation platform is a data receiver.

[0032] The business data generated in the operation of different hotel function systems is different, for example, the occupancy rate of PMS, the cancellation amount of CRS, the transaction success rate of POS, etc.

[0033] The data aggregation platform acquires the function data corresponding to each of the plurality of hotel function systems, including: According to the preset demand level of the function data of each hotel function system, a preset mode of the data aggregation platform acquiring the function data from each hotel function system is determined. The preset demand level is determined according to the data change frequency, the real-time requirement, and the data amount.

[0034] The high and low of the data change frequency can be determined according to the actual demand, and specifically, the preset change frequency threshold can be set to determine whether the data change frequency is high or low. The real-time requirement is determined based on the degree of allowed delay, if the degree of allowed delay is high, then the real-time requirement is low, if the degree of allowed delay is low, then the real-time requirement is high. The size of the data amount is determined according to the ability of the system to ingest data at one time, if it exceeds 70% of the ability of the system to ingest data at one time, it is considered that the data amount is large, otherwise it is considered that the data amount is small.

[0035] In particular, the preset requirement level in the embodiments of the present application mainly includes a high level and a low level. The high level has higher requirements, that is, a higher data change frequency, a lower degree of allowed delay, and a relatively small data volume. The low level has relatively lower requirements, that is, a lower data change frequency, a higher degree of allowed delay, and a relatively large data volume. The preset requirement level corresponds to a preset mode one by one, and the preset mode includes that a data sender actively sends data to a data receiver and that the data receiver actively acquires data from the data sender.

[0036] If the preset requirement level of the hotel function system is the high level, the preset mode corresponding to the hotel function system is that the hotel function system actively sends function data to the state monitoring system according to a first preset frequency. If the preset requirement level of the hotel function system is the low level, the state monitoring system actively acquires function data from the hotel function system according to a second preset frequency. The first preset frequency and the second preset frequency can be selected according to actual requirements, and the embodiments of the present application do not limit this.

[0037] In the actual scene of the hotel system application, the mode of the data receiver actively acquiring data from the data sender can be used under the following conditions: infrequent data update, no large number of invalid requests caused by pulling (such as static configuration and daily report), allowed delay (such as pulling logs by a timing task and batch data synchronization), the data receiver only needs part of the data (such as page query and condition filtering), and the data receiver can be in a weak network environment (such as a mobile terminal) and the push cost is high.

[0038] In the actual scene of the hotel system application, the mode of the data sender actively sending data to the data receiver can be used under the following conditions: avoiding delay caused by polling (such as stock prices, chat messages, and Internet of Things sensor data), triggering push only when there is a change to save bandwidth and computing resources (such as event notification), and the receiver can timely process the push data (such as a high-concurrency server).

[0039] Further, the data aggregation platform acquires function data corresponding to each of the plurality of hotel function systems, and stores the function data of the plurality of hotel function systems, including: The data aggregation platform acquires original function data corresponding to each of the plurality of hotel function systems, and converts the original function data of each of the hotel function systems into benchmark function data meeting a preset benchmark data format requirement; and stores the benchmark function data of each of the hotel function systems.

[0040] The original function data generated by each hotel function system can be in different data formats, such as SQL for PMS, NoSQL for door lock system, and binary log for POS. Therefore, a unified data model standard (such as Apache Avro or JSON Schema) can be defined, that is, a preset baseline data format requirement is defined. Then, the middleware (such as Apache Kafka) is used to receive data in different protocols from the hotel function systems in different data formats, and then the data in different data formats is converted into data in a unified format according to the preset baseline data format requirement. Finally, the baseline function data of each hotel function system is stored.

[0041] Regarding step S22, the state monitoring system determines whether there is a target system with state anomaly in the plurality of hotel function systems based on the various function data stored by the data aggregation platform.

[0042] The state monitoring system includes a plurality of pre-trained machine learning models, and each machine learning model corresponds to one of the hotel function systems.

[0043] The state monitoring system determines whether there is a target system with state anomaly in the plurality of hotel function systems based on the various function data stored by the data aggregation platform, including: For each hotel function system, the state monitoring system inputs the function data of the hotel function system in the data aggregation platform within a target historical period into the machine learning model of the hotel function system, so that the machine learning model outputs a monitoring result representing whether the hotel function system is state abnormal; wherein the machine learning model is trained based on the state normal function data in a period before the target historical period. According to the monitoring result of each hotel function system, it is determined whether there is a target system with state anomaly in the plurality of hotel function systems.

[0044] Embodiments of the present application deploy a dedicated machine learning model for each hotel function system (such as PMS, CRS, POS, etc.); train the model with historical normal data to establish a baseline behavior pattern; achieve anomaly detection by comparing the deviation of current data from the baseline pattern in real time; and locate the target abnormal system based on a multi-system collaborative judgment mechanism.

[0045] The machine learning model corresponding to each hotel function system is trained based on the state normal function data in a period before the target historical period. Each machine learning model can be updated at a certain period to ensure that the machine learning model matches the recent actual operation of the hotel, so as to improve the accuracy of the machine learning model.

[0046] Further, the machine learning model of each hotel function system can include two different models, one of which can be trained based on historical data associated with weekend time periods and normal, which is denoted as a weekend model in the embodiments of the present application. The other model can be trained based on historical data associated with working time periods and normal, which is denoted as a weekday model in the embodiments of the present application.

[0047] The target historical time period in the embodiments of the present application can be a time window [T-Δt, T] of a preset time length Δt (such as 72 hours) before the current detection time point T. Further, the target historical time period in the embodiments of the present application can also be the latest weekend time period before the current detection time point. The weekend time period can include Friday to Sunday. Alternatively, it can also be the latest working day time period before the current detection time point, and the working day time period can include Monday to Thursday.

[0048] By inputting the function data of each hotel function system in the latest weekend time period into the weekend model, whether each hotel function system has a state anomaly in the current weekend time period can be obtained. Since the weekend model is trained based on system data of normal weekend in the historical stage, it means that the embodiments of the present application horizontally compare the system state of the latest weekend time period with that of the weekend time period before it to determine whether the hotel function system is in a normal state in the weekend time period.

[0049] Similarly, by inputting the function data of each hotel function system in the latest working day time period into the weekday model, whether each hotel function system has a state anomaly in the current working day time period can be obtained. Since the weekday model is trained based on system data of normal working day in the historical stage, it means that the embodiments of the present application horizontally compare the system state of the latest working day time period with that of the working day time period before it to determine whether the hotel function system is in a normal state in the working day time period.

[0050] It can be seen that the embodiments of the present application realize the same period and period-on-period analysis, that is, comparison with historical same period data to determine whether each hotel function system is in a normal state, which can be more in line with the actual operation and management state of the hotel in different time periods to improve the judgment accuracy of the machine learning model.

[0051] Further, there can be a group of function systems in each hotel function system whose data are closely related. For example, when a guest orders food, the POS records the order, the PMS records the room account, and the PMS settles the account when the guest checks out, which makes the data of the POS and the PMS closely related. For another example, when the BAS detects that there is no one in the public area and the outdoor temperature is less than 26°C, the lobby air conditioner needs to be turned off and the RCU needs to be notified to raise the upper limit of the room temperature to 26°C, which makes the data of the RCU and the BAS closely related. In addition, when the PMS room status is not updated in time and the RCU supplies power to a room that has been checked out, it leads to energy waste, which reflects the close relationship between the data of the PMS and the RCU.

[0052] In an embodiment of the present application, at least two hotel function systems in which the function data of the multiple hotel function systems are related to each other are recorded as an associated function system group. Each hotel function system can exist in different associated function system groups, and each associated function system group includes at least two different hotel function systems. Which hotel function systems are included in each associated function system group can be adjusted and optimized according to actual needs.

[0053] The state monitoring system determines whether there is a target system with a state anomaly in the multiple hotel function systems based on the various function data stored in the data aggregation platform, including: For each hotel function system, the state monitoring system obtains an independent monitoring result representing whether the hotel function system has a state anomaly based on the function data of the hotel function system in the data aggregation platform; For each associated function system group, the state monitoring system obtains an associated monitoring result representing whether the associated function system group has a state anomaly based on the function data of each hotel function system in the associated function system group in the data aggregation platform; According to the independent monitoring result and the associated monitoring result, it is determined whether there is a target system with a state anomaly in the multiple hotel function systems; the target system includes the hotel function system and / or the associated function system group.

[0054] For the determination of the independent monitoring result, reference can be made to the foregoing content, including: for each hotel function system, the state monitoring system inputs the function data of the hotel function system in the data aggregation platform in the target historical period into the machine learning model of the hotel function system, so that the machine learning model outputs a monitoring result representing whether the hotel function system has a state anomaly; wherein the machine learning model is trained based on the function data of the state normal in the period before the target historical period.

[0055] As to the determination of the correlation monitoring result, the correlation learning model can be trained in advance according to historical data of all hotel function systems included in the correlation function system group. It should be noted that the historical data refers to normal historical data before the target historical period and includes historical data of multiple different hotel function systems. The function data of each hotel function system in the correlation function system group in the target historical period is input into the correlation learning model to obtain the correlation monitoring result corresponding to the point of each correlation function system group.

[0056] Based on the independent monitoring result and the correlation monitoring result, the embodiments of the present application can determine which hotel function system has an anomaly in the target historical period and which correlation function system group has an anomaly. The embodiments of the present application compare each hotel function system in the time sequence in the longitudinal direction to determine whether the state of a single hotel function system is normal, and compare several associated hotel function systems in the horizontal direction and in the longitudinal direction in the time sequence to determine whether the state of each correlation function system group is normal. It can be seen that the scheme provided by the embodiments of the present application can cover the state monitoring of all hotel function systems of a hotel, realize global anomaly perception and accurate root cause positioning, and effectively solve the three technical bottlenecks existing in the traditional monitoring: (1) Breakthrough of single-point monitoring blind area Through the dual-dimensional fusion of longitudinal time sequence analysis (single system historical behavior modeling) and horizontal correlation analysis (multi-system business logic verification), the problem of missing detection of correlation anomalies by isolated system monitoring is overcome. For example, it is detected that the transaction volume of the POS system is normal alone, but deviates from the PMS account rate (horizontal correlation anomaly); it is found that the CRS reservation volume is within the historical fluctuation range, but its week-on-week drops by 50% (longitudinal time sequence anomaly).

[0057] (2) Complex anomaly pattern recognition Using the modeling capability of machine learning for non-linear relationships, implicit faults that cannot be covered by traditional threshold alarms are captured.

[0058] (3) Fault propagation chain blocking Based on the root cause positioning technology of dynamic topology, accurate isolation and active defense are realized when the correlation system group is abnormal. Accurate isolation refers to distinguishing the source system (such as CRS price synchronization failure) and the affected system (such as PMS revenue calculation distortion). Active defense refers to automatically triggering the CRS reservation channel to close when it is detected that the PMS-CRS room state difference continues to expand.

[0059] In the case that the target system has a state abnormality in the plurality of hotel function systems, the state monitoring system obtains, from the data aggregation platform, to-be-analyzed data related to the target system, determines an abnormal source causing the state abnormality of the target system based on the to-be-analyzed data, and outputs abnormal alarm information according to the abnormal source.

[0060] The to-be-analyzed data includes function data of the target system in different time periods, function data of other systems associated with the target system, and other data associated with the target system. The function data of the target system in different time periods refers to the function data corresponding to the target system itself, and specifically can include data in a target historical period and data before the target historical period. The function data of other systems associated with the target system mainly refers to function data of each hotel function system in the associated function system group where the target system is located. The other data associated with the target system mainly refers to related data such as weather, temperature, electricity price fluctuation, and equipment maintenance cost.

[0061] The determination of the abnormal source causing the state abnormality of the target system based on the to-be-analyzed data includes: inputting the to-be-analyzed data into a pre-trained root cause probability calculation model to output a root cause probability result, wherein the root cause probability result includes at least one abnormal source causing the state abnormality of the target system and an actual probability of each abnormal source causing the state abnormality of the target system.

[0062] Historical abnormal events are collected, each event can include an abnormal index of a hotel function system (such as PMS occupancy rate), associated system state data (such as CRS synchronization log, network delay), environmental context (such as holiday label, weather coding, etc.), and root cause probability of each historical abnormal event. Then, these data are used to train an original machine learning model to obtain a root cause probability calculation model.

[0063] The output of the abnormal alarm information according to the abnormal source includes: determining an abnormal alarm level according to a device type and an influence range corresponding to the abnormal source; and outputting the abnormal alarm information by using a preset alarm mode matched with the abnormal alarm level.

[0064] For example, the PMS-CRS correlation group represents room state consistency and price synchronization delay, and the detection frequency is real-time. If the correlation group has an exception, the exception alarm level belongs to the disaster level (representing the most serious level). The RCU-BAS correlation group represents energy consumption deviation rate and instruction response time, which can be polled every 5 minutes. If the correlation group has an exception, the exception alarm level belongs to the serious level (representing the second most serious level). The PMS-POS correlation group represents account failure rate and transaction reconciliation difference, which can be updated every hour. If the correlation group has an exception, the exception alarm level belongs to the high-risk level (representing the third most serious level). The CRS-RMS correlation group represents price strategy execution rate and reservation conversion rate, which can update data every 30 minutes. If the correlation group has an exception, the exception alarm level belongs to the high-risk level.

[0065] According to different severity levels, that is, according to different exception alarm levels, different preset alarm modes are selected. For example, the higher the exception alarm level, the higher the management level of the alarm object corresponding to the preset alarm mode, which can be implemented by means of short message, telephone, and network communication tool.

[0066] In summary, the embodiment of the present application obtains the respective function data of the plurality of hotel function systems based on the data aggregation platform, and stores the function data of the plurality of hotel function systems. The state monitoring system determines whether there is a target system with a state exception in the plurality of hotel function systems based on the various function data stored by the data aggregation platform. In the case where there is a target system with a state exception in the plurality of hotel function systems, the state monitoring system obtains the to-be-analyzed data related to the target system from the data aggregation platform, determines the exception source causing the state exception of the target system based on the to-be-analyzed data, and outputs exception alarm information according to the exception source. It can be seen that the embodiment of the present application can cope with the state monitoring of all management and control systems of the hotel. The state monitoring system identifies system exceptions in real time, and compared with the traditional threshold alarm, the average response time can be shortened to seconds, and the response speed is improved. Through intelligent analysis of the model, the exception identification is more accurate, and the false alarm rate is significantly reduced. With the help of the call chain and the dependency topology, the root cause positioning accuracy can reach the module / interface level, and the fault handling efficiency is improved. The closed-loop process of automatic identification, positioning, and notification can significantly reduce manual participation and improve the degree of operation and maintenance automation. By discovering potential problems in advance and responding in a timely manner, the overall system failure risk is reduced, and the system stability is enhanced.

[0067] Based on the same inventive concept, the embodiments of the present application provide a system as shown in Figure 3A hotel data monitoring device is shown, the hotel includes a state monitoring system, a data aggregation platform and a plurality of hotel function systems, the state monitoring system and the data aggregation platform interact with information, the data aggregation platform and a plurality of the hotel function systems interact with information; the device comprises: The data acquisition and storage module 31 is used for the data aggregation platform to acquire the respective corresponding function data of a plurality of the hotel function systems, and to store the function data of a plurality of the hotel function systems; The target system determination module 32 is used for the state monitoring system to determine whether there is a target system with state abnormalities in a plurality of the hotel function systems based on various function data stored by the data aggregation platform; The abnormal response module 33 is used for the state monitoring system to obtain the target system related to the target system from the data aggregation platform in the case of a plurality of the hotel function systems with state abnormalities, to determine the abnormal source causing the state abnormalities of the target system based on the analysis data, and to output abnormal alarm information according to the abnormal source.

[0068] Further, the data acquisition and storage module 31 is used for: According to the preset demand level of the function data of each hotel function system, the preset mode of the data aggregation platform acquiring the function data from each hotel function system is determined; the preset demand level is determined according to the data change frequency, real-time requirement and data volume; If the preset demand level of the hotel function system is high, the preset mode corresponding to the hotel function system is that the hotel function system actively sends function data to the state monitoring system according to the first preset frequency; if the preset demand level of the hotel function system is low, the state monitoring system actively collects function data from the hotel function system according to the second preset frequency.

[0069] Further, the data acquisition and storage module 31 is used for: The data aggregation platform acquires the respective corresponding original function data of a plurality of the hotel function systems, and converts the original function data of each hotel function system into benchmark function data meeting the preset benchmark data format requirement; The benchmark function data of each hotel function system is stored.

[0070] Further, the state monitoring system includes a plurality of machine learning models pre-trained, and the machine learning models correspond to the hotel function systems one by one; The target system determination module 32 is used for: For each of the hotel function systems, the state monitoring system inputs the function data of the hotel function system in the data aggregation platform in a target historical period into the machine learning model of the hotel function system, so that the machine learning model outputs a monitoring result representing whether the hotel function system is state abnormal; wherein the machine learning model is trained based on state-normal function data in a period before the target historical period; According to the monitoring results of each of the hotel function systems, it is determined whether there is a target system that is state abnormal in the plurality of hotel function systems.

[0071] Further, at least two of the plurality of hotel function systems, in which function data exist mutual influences, are recorded as an associated function system group; The target system determination module 32 is configured to: For each of the hotel function systems, the state monitoring system obtains an independent monitoring result representing whether the hotel function system is state abnormal based on the function data of the hotel function system in the data aggregation platform; For each of the associated function system groups, the state monitoring system obtains an associated monitoring result representing whether the associated function system group is state abnormal based on the function data of each of the hotel function systems in the associated function system group in the data aggregation platform; According to the independent monitoring result and the associated monitoring result, it is determined whether there is a target system that is state abnormal in the plurality of hotel function systems; the target system includes the hotel function system and / or the associated function system group.

[0072] Further, the to-be-analyzed data includes function data of the target system in different periods, function data of other systems associated with the target system, and other data associated with the target system; The abnormal response module 33 is configured to: Input the to-be-analyzed data into a pre-trained root cause probability calculation model to output a root cause probability result, wherein the root cause probability result includes at least one abnormal source causing the state abnormal of the target system and an actual probability of each abnormal source causing the state abnormal of the target system.

[0073] Further, the abnormal response module 33 is configured to: Determine an abnormal alarm level according to the device type and the influence range corresponding to the abnormal source; Output the abnormal alarm information by using a preset alarm mode matched with the abnormal alarm level.

[0074] Based on the same inventive concept, the embodiments of the present application provide a system for analyzing a target system that is state abnormal, as shown inFigure 4 An electronic device is shown, comprising: a processor 41; a memory 42 for storing instructions executable by the processor 41; wherein the processor 41 is configured to perform to implement a hotel data monitoring method as provided in the foregoing.

[0075] Based on the same inventive concept, the embodiments of the present application provide a non-transitory computer-readable storage medium, when instructions in the storage medium are executed by a processor 41 of an electronic device, the electronic device is enabled to perform a hotel data monitoring method as provided in the foregoing.

[0076] Since the electronic device introduced in the embodiments is the electronic device used to implement the method of information processing in the embodiments of the present application, the specific implementation of the electronic device of the embodiments and its various forms of changes can be understood by those skilled in the art based on the method of information processing introduced in the embodiments of the present application, so the electronic device how to implement the method in the embodiments of the present application will not be introduced in detail here. As long as the electronic device used to implement the method of information processing in the embodiments of the present application is implemented by those skilled in the art, it belongs to the scope of the present application.

[0077] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage, etc.) containing computer-usable program code.

[0078] The present application is described with reference to flowcharts and / or block diagrams of the method, device (system), and computer program product according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to produce a machine, so that the instructions executed by the computer or other programmable data processing devices produce a device that implements the functions specified in the flowcharts and / or block diagrams. Figure 1 The function specified in one flow or multiple flows and / or blocks Figure 1 The device that implements the function specified in one block or multiple blocks.

[0079] These computer program instructions can also be stored in a computer- readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the Figure 1 function specified in the flow or flows and / or blocks Figure 1 of the block or blocks.

[0080] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the Figure 1 function specified in the flow or flows and / or blocks Figure 1 of the block or blocks.

[0081] Although preferred embodiments of the application have been described herein, substitutions and alterations are possible in view of the teachings of this application. Accordingly, the appended claims are intended to encompass all such substitutions and alterations. It will be apparent to those skilled in the art that various modifications and variations can be made to the present application without departing from the spirit or scope of the application. Thus, it is intended that the present application cover the modifications and variations of this application provided they come within the scope of the appended claims and their equivalents.

[0082] It will be apparent to those skilled in the art that various modifications and variations can be made to the present application without departing from the spirit or scope of the application. Thus, it is intended that the present application cover the modifications and variations of this application provided they come within the scope of the appended claims and their equivalents.

Claims

1. A hotel data monitoring method, characterized in that, The hotel includes a status monitoring system, a data aggregation platform, and multiple hotel functional systems. The status monitoring system interacts with the data aggregation platform, and the data aggregation platform interacts with the multiple hotel functional systems. The method includes: The data aggregation platform acquires functional data corresponding to each of the multiple hotel functional systems and stores the functional data of the multiple hotel functional systems. The status monitoring system determines whether there are target systems with abnormal statuses among the multiple hotel functional systems based on various functional data stored in the data aggregation platform. In the event that the target system is in an abnormal state among multiple hotel functional systems, the status monitoring system obtains data to be analyzed related to the target system from the data aggregation platform, determines the abnormal source causing the abnormal state of the target system based on the data to be analyzed, and outputs abnormal alarm information according to the abnormal source.

2. The hotel data monitoring method as described in claim 1, characterized in that, The data aggregation platform acquires functional data corresponding to each of the multiple hotel functional systems, including: Based on the preset demand levels of the functional data of each of the hotel's functional systems, the preset method for the data aggregation platform to obtain functional data from each of the hotel's functional systems is determined; the preset demand levels are determined based on the data change frequency, real-time requirements, and data volume. Specifically, if the preset demand level of the hotel functional system is high, then the preset method corresponding to the hotel functional system is that the hotel functional system actively sends functional data to the status monitoring system at a first preset frequency; if the preset demand level of the hotel functional system is low, then the status monitoring system actively collects functional data from the hotel functional system at a second preset frequency.

3. The hotel data monitoring method as described in claim 1, characterized in that, The data aggregation platform acquires functional data corresponding to each of the multiple hotel functional systems and stores the functional data of the multiple hotel functional systems, including: The data aggregation platform acquires the original functional data corresponding to each of the multiple hotel functional systems, and converts the original functional data of each hotel functional system into benchmark functional data that meets the requirements of the preset benchmark data format. The baseline functional data of each of the aforementioned hotel functional systems are stored.

4. The hotel data monitoring method as described in claim 1, characterized in that, The status monitoring system includes multiple pre-trained machine learning models, and each machine learning model corresponds one-to-one with the hotel's functional system. The status monitoring system, based on various functional data stored on the data aggregation platform, determines whether there are target systems with abnormal statuses among multiple hotel functional systems, including: For each of the hotel functional systems, the status monitoring system inputs the functional data of the hotel functional system within the target historical period from the data aggregation platform into the machine learning model of the hotel functional system, so that the machine learning model outputs a monitoring result characterizing whether the hotel functional system is in an abnormal state; wherein, the machine learning model is trained based on functional data of normal state in the period prior to the target historical period; Based on the monitoring results of each of the hotel functional systems, determine whether there is a target system with an abnormal status among the multiple hotel functional systems.

5. A hotel data monitoring method as described in claim 1, characterized in that, At least two of the hotel functional systems in which the functional data of the multiple hotel functional systems are mutually influential are referred to as the related functional system group; The status monitoring system, based on various functional data stored on the data aggregation platform, determines whether there are target systems with abnormal statuses among multiple hotel functional systems, including: For each of the hotel functional systems, the status monitoring system obtains an independent monitoring result characterizing whether the hotel functional system is in an abnormal state based on the functional data of the hotel functional system in the data aggregation platform. For each of the associated functional system groups, the status monitoring system obtains an associated monitoring result characterizing whether the associated functional system group is in an abnormal state based on the functional data of each hotel functional system in the associated functional system group in the data aggregation platform. Based on the independent monitoring results and the associated monitoring results, determine whether there is a target system with an abnormal status among the multiple hotel functional systems; the target system includes the hotel functional system and / or the associated functional system group.

6. The hotel data monitoring method as described in claim 1, characterized in that, The data to be analyzed includes the functional data of the target system at different time periods, the functional data of other systems associated with the target system, and other data associated with the target system; The process of determining the anomaly source causing the abnormal state of the target system based on the data to be analyzed includes: The data to be analyzed is input into a pre-trained root cause probability calculation model, and the root cause probability results are output. The root cause probability results include at least one abnormal source that causes the state abnormality of the target system and the actual probability of each abnormal source causing the state abnormality of the target system.

7. A hotel data monitoring method as described in claim 1, characterized in that, The step of outputting abnormal alarm information based on the abnormal source includes: The alarm level is determined based on the device type and the scope of impact corresponding to the anomaly source. The abnormal alarm information is output using a preset alarm method that matches the abnormal alarm level.

8. A hotel data monitoring device, characterized in that, The hotel includes a status monitoring system, a data aggregation platform, and multiple hotel functional systems. The status monitoring system interacts with the data aggregation platform, and the data aggregation platform interacts with the multiple hotel functional systems. The device includes: The data acquisition and storage module is used by the data aggregation platform to acquire functional data corresponding to each of the multiple hotel functional systems and to store the functional data of the multiple hotel functional systems. The target system determination module is used by the status monitoring system to determine whether there is a target system with an abnormal status among the multiple hotel functional systems based on various functional data stored in the data aggregation platform. An anomaly response module is used to, in the event that the target system has an abnormal status among multiple hotel functional systems, obtain data to be analyzed related to the target system from the data aggregation platform, determine the anomaly source causing the abnormal status of the target system based on the data to be analyzed, and output anomaly alarm information according to the anomaly source.

9. An electronic device, characterized in that, include: processor; Memory used to store the processor's executable instructions; The processor is configured to execute a hotel data monitoring method as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium, wherein instructions in the storage medium, when executed by a processor of an electronic device, enable the electronic device to perform a hotel data monitoring method as described in any one of claims 1 to 7.

Images