Hotel elevator pre-scheduling method and device based on user portrait, equipment and medium

Abstract

This invention provides a hotel elevator pre-scheduling method, apparatus, equipment, and medium based on user profiles, relating to the field of elevator scheduling technology. The method includes: constructing a basic profile and a real-time iterative profile of the target guest based on guest attribute data; real-time monitoring of the target guest's room door time-series behavior data, distinguishing the target guest's stay status and travel scenarios based on the room door time-series behavior data and the basic profile, and calculating a travel intention confidence level based on the travel scenario and the real-time iterative profile; if the travel intention confidence level is greater than a preset confidence threshold, calculating the predicted time for the elevator to reach the target guest's floor based on the elevator's real-time operating status; and determining the timing of issuing an elevator call command based on the matching relationship between walking speed in the real-time iterative profile and the predicted time, so that the elevator arrives at the target guest's floor in advance. This invention transforms elevator scheduling from a passive response to proactive scheduling, improving guest experience and optimizing elevator operating efficiency.

Description

Technical Field

[0001] This invention relates to the field of elevator scheduling technology, and in particular to a method, device, equipment and medium for pre-scheduling hotel elevators based on user profiles. Background Technology

[0002] In hotel operations, elevator scheduling directly impacts the guest experience and the hotel's operational efficiency.

[0003] Existing hotel elevator systems typically employ a passive response mode, where guests press the call button upon arriving at the elevator lobby, and the elevator control system responds based on the current operating status and existing requests. This mode has significant drawbacks: guests need time to walk from their rooms to the elevator lobby, and when the elevator is either departing or busy, guests often face long waits, especially during peak check-out times or after meetings. Simultaneous calls from multiple floors can lead to frequent elevator stops, capacity strain, and significantly extended wait times. Some hotels have attempted to optimize elevator allocation by installing destination selection terminals in the lobby or using group control algorithms, but these solutions still rely on guests actively triggering calls and cannot anticipate their travel intentions, thus still resulting in long wait times and low elevator capacity utilization. Summary of the Invention

[0004] This invention provides a hotel elevator pre-scheduling method, device, equipment, and medium based on user profiles to solve the problem that existing elevator scheduling systems rely on passive triggering, resulting in long waiting times for guests and low elevator capacity utilization.

[0005] In a first aspect, embodiments of the present invention provide a hotel elevator pre-scheduling method based on user profiles, including: Based on the guest attribute data of the target guests, construct a basic profile and a real-time iterative profile of the target guests; Real-time monitoring of guest room door time-series behavior data of target guests; differentiation of guest room door time-series behavior data and basic profiles to distinguish the stay status and travel scenarios of target guests; and calculation of travel intention confidence based on travel scenarios and real-time iterative profiles. If the confidence level of the travel intention is greater than the preset confidence level threshold, the predicted time for the elevator to reach the floor where the target resident is located is calculated based on the real-time operating status of the elevator. Based on the matching relationship between walking speed and predicted time in the real-time iterative profile, the timing of issuing elevator call commands is determined so that the elevator can arrive at the floor where the target resident is located in advance.

[0006] In one possible implementation, guest attribute data includes room floor, room number, distance from the room to the elevator lobby, historical check-in behavior, current check-out time, and the target floor corresponding to the reserved service. Based on the guest attribute data of the target guest, a basic profile and a real-time iterative profile of the target guest are constructed, including: The basic profile of the target guest is obtained by combining the target guest's room floor, room number, distance from the room to the elevator lobby, historical check-in behavior, current check-out time, and the target floor corresponding to the reservation service. The travel time points are extracted from the historical room door time sequence behavior data of the target guests, and cluster analysis is performed to obtain the high-frequency travel time periods of the target guests; Based on the statistical frequency distribution of the target guests' historical travel types, the travel type preferences of the target guests are obtained; The target floor preference of the target guests is obtained by statistically analyzing the frequency distribution of their historical travel target floor records; Extract the ratio of displacement to time from the historical movement trajectory points of the target guest, and calculate the average value to obtain the walking speed of the target guest; The probability of trip cancellation for the target guest is obtained by calculating the sliding window ratio from historical travel attempt and cancellation records; By combining high-frequency travel times, travel type preferences, target floor preferences, walking speed, and travel cancellation probability, a real-time iterative profile of the target guest is obtained.

[0007] In one possible implementation, the guest room door time-series behavior data includes door opening and closing events; real-time monitoring of the target guest's guest room door time-series behavior data distinguishes between the target guest's stay status and travel scenario, including: Mark the first time a target guest swipes their card to enter the room after check-in as the check-in entry event and use it as the starting point of the stay period; During the stay period, if a door opening event is detected and no further door opening events occur after a preset silent window period, the door opening event is determined to be an outgoing door opening, and the stay is switched to non-stay status. If the current time is less than the time of this check-out, and the guest room door timing behavior data includes the door opening mode within the second preset threshold time after the door is closed, then it is determined to be a check-out travel scenario. If the current time is less than the time of any scheduled service and an opening event occurs in the guest room door time sequence behavior data, it is determined to be a fixed service travel scenario. If a door opening event occurs in the guest room door sequence behavior data, but does not meet the criteria for check-out or fixed service travel scenarios, it is determined to be a temporary travel scenario.

[0008] In one possible implementation, the confidence level of travel intention is calculated based on travel scenarios and real-time iterative profiles, including: If the scenario is determined to be a check-out travel scenario, the travel confidence score will be assigned the first preset score. If the scenario is determined to be a fixed-service travel scenario, the travel confidence score is assigned a second preset score; wherein the second preset score is less than the first preset score. If the scenario is determined to be a temporary travel scenario, the travel confidence score is obtained by weighted summation of the confidence score influencing factors. Among them, the confidence score influencing factors include at least one of the following: room door opening and closing behavior factor, historical travel frequency factor, and travel cancellation penalty factor.

[0009] In one possible implementation, the feature is that, after issuing the elevator call command, it further includes: If a target resident is detected entering the elevator within the fourth preset threshold time after the elevator arrives, and the elevator floor change matches the current travel scenario, then it is determined to be a valid dispatch. The walking speed in the real-time iterative profile is updated using an exponentially weighted moving average method. And / or, use event counting to update the high-frequency travel time periods and target floor preferences in the real-time iterative profile.

[0010] In one possible implementation, the feature is that, after issuing the elevator call command, it further includes: If no target guest is detected entering the elevator within the fourth preset threshold time after the elevator arrives, and the room door status shows that the door is closed again, it is determined to be an invalid dispatch. Cancel the issued elevator call command, reduce the preset confidence threshold by a preset percentage, and increase the probability of trip cancellation in the real-time iterative profile by a preset increment.

[0011] In one possible implementation, the method further includes: When there are multiple target guests and the current time is the peak check-out period or the peak time for meeting end, multiple elevator call requests on the same floor will be combined and processed, and elevator capacity will be allocated according to the principle of prioritizing check-out guests over regular guests and VIP members over regular members.

[0012] Secondly, embodiments of the present invention provide a hotel elevator pre-scheduling device based on user profiles, comprising: The profile building module is used to build a basic profile and a real-time iterative profile of the target guest based on the guest attribute data of the target guest; The travel judgment module is used to monitor the time-series behavior data of target guests at the guest room door in real time. Based on the time-series behavior data of the guest room door and the basic profile, it distinguishes the stay status and travel scenario of the target guests, and calculates the confidence level of travel intention based on the travel scenario and the real-time iterative profile. The time prediction module is used to calculate the predicted time for the elevator to reach the floor where the target guest is located based on the real-time operation status of the elevator when the confidence level of the travel intention is greater than the preset confidence level threshold. The instruction issuance module is used to determine the timing of issuing elevator call instructions based on the matching relationship between walking speed and predicted time in the real-time iterative profile, so that the elevator can arrive at the floor where the target resident is located in advance.

[0013] Thirdly, embodiments of the present invention provide an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method described in the first aspect or any possible implementation thereof.

[0014] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in the first aspect or any possible implementation thereof.

[0015] The hotel elevator pre-scheduling method, apparatus, equipment, and medium based on user profiles provided in this invention construct a basic profile and a real-time iterative profile of the target guest, and monitor the time-series behavior data of guest room doors in real time. By combining information such as guest door opening and closing actions, reservation service information, and behavioral preferences, the system proactively distinguishes the guest's stay status and travel scenario, and then calculates the travel intention confidence level. When the travel intention confidence level meets the standard, the system estimates the predicted elevator arrival time based on the real-time elevator operating status and determines the timing of issuing the elevator call command based on the travel characteristics in the real-time iterative profile, so that the elevator arrives at the guest's floor ahead of time. This method transforms elevator scheduling from a passive response to proactive pre-scheduling, initiating the scheduling process before the guest leaves the room, achieving synchronous arrival of the elevator and the guest, and significantly reducing the guest's waiting time in the elevator lobby. Simultaneously, since the scheduling decision is based on profiles and door behavior, it can effectively filter out false triggers, avoiding the waste of elevator resources due to ineffective scheduling, thereby improving the guest experience and optimizing elevator operating efficiency. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the implementation of the hotel elevator pre-scheduling method based on user profiles provided in this embodiment of the invention. Figure 2 This is a schematic diagram of the structure of the hotel elevator pre-scheduling device based on user profile provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0017] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0018] See Figure 1 The document illustrates a flowchart of the implementation of the hotel elevator pre-scheduling method based on user profiles provided in an embodiment of the present invention, which is described in detail below: Step 101: Construct a basic profile and a real-time iterative profile of the target guest based on the guest attribute data of the target guest.

[0019] In this embodiment, it is necessary to construct a basic profile and a real-time iterative profile of the target guest based on the guest attribute data.

[0020] Guest attribute data typically comes from the hotel management system, including various information entered by guests during check-in and their historical behavior records. This data serves as the raw material for subsequent profile construction. The basic profile focuses on the guest's static or quasi-static characteristics, while the real-time iterative profile reflects the guest's dynamically changing behavioral habits during their stay.

[0021] A basic profile can be understood as the initial set of features obtained when a guest checks in. It doesn't change rapidly with short-term guest behavior and is mainly used for judgment during the cold start phase. A real-time iterative profile, on the other hand, is updated based on actual behavioral data after each guest's trip, thus continuously approximating the guest's true behavioral patterns. For example, when a guest first checks in, the system hasn't yet acquired their walking speed data or target floor preference; this information needs to be gradually accumulated and corrected during their stay.

[0022] Therefore, the basic profile and the real-time iterative profile exist in parallel, independent of each other but jointly serving the judgment of travel intention.

[0023] Step 102: Monitor the time-series behavior data of the target guests at the guest room door in real time, distinguish the stay status and travel scenarios of the target guests based on the time-series behavior data of the guest room door and the basic profile, and calculate the confidence level of travel intention based on the travel scenario and the real-time iterative profile.

[0024] In this embodiment, the guest room door timing behavior data includes door opening events, door closing events, and a precise timestamp for each event, which is typically collected and reported in real time by sensors installed on the door lock or door magnet.

[0025] The significance of distinguishing stay status lies in determining whether a guest is currently in the room. Only guests within their stay period can open the door to represent a new travel intention; if the guest has already left, subsequent door openings may indicate a return and should not trigger pre-scheduling repeatedly. Stay status can be simply understood as the time period from when a guest first swipes their card to enter the room until it is confirmed that the guest has left.

[0026] The travel scenario further refines the types of travel purposes, such as checking out, dining, attending meetings, or going out temporarily. The urgency and predictability of guests' travel vary in different scenarios, so the methods for calculating the confidence level of travel intention will also differ.

[0027] The travel intention confidence score is a quantitative indicator used to represent the system's degree of certainty that a guest will soon use the elevator. The score ranges from 0 to 100, with higher scores indicating greater confidence. This confidence score is calculated by combining current door behavior patterns, check-out times and reservation service information from the basic profile, and historical travel patterns from the real-time iterative profile.

[0028] Step 103: If the confidence level of the travel intention is greater than the preset confidence level threshold, the predicted time for the elevator to reach the floor where the target guest is located is calculated based on the real-time operating status of the elevator.

[0029] In this embodiment, the confidence threshold needs to be set in a balance between scheduling success rate and resource waste. If the threshold is too low, frequent invalid scheduling may occur due to misjudgment, consuming elevator resources; if the threshold is too high, scheduling opportunities may be missed, causing passengers to wait. Through engineering practice, setting the threshold to 60 points can achieve a better overall effect.

[0030] Once the confidence level is met, the system needs to quickly estimate the time required for the elevator to travel from its current location to the guest's floor. This predicted time is not a fixed value but dynamically changes, relying on real-time data from the elevator control system, including the elevator's current floor, direction of travel (up or down), registered floor button requests inside the car, and existing call requests on the external call boxes at each floor. Each elevator typically requires a fixed travel time to traverse one floor, such as 3 seconds, while each stop, door opening, and closing also incurs a fixed time cost, such as 2 seconds. By accumulating the travel time and possible stop times of all necessary floors along the path from the current location to the target floor, the approximate arrival time of the elevator can be estimated.

[0031] Step 104: Based on the matching relationship between walking speed and predicted time in the real-time iterative profile, determine the timing of issuing the elevator call command so that the elevator can arrive at the floor where the target resident is located in advance.

[0032] In this embodiment, the walking time for guests from their guest rooms to the elevator lobby is not fixed, and different guests walk at different speeds. Even for the same guest, the walking speed may vary at different times or under different physical conditions. The walking speed in the real-time iterative profile is a personalized parameter obtained by collecting the guest's historical walking trajectory points, calculating the displacement-to-time ratio, and taking the moving average.

[0033] By combining the distance from the guest room to the elevator lobby stored in the basic profile, the estimated travel time for the guest can be estimated. The system compares the predicted elevator arrival time with the predicted guest travel time: ideally, the elevator will arrive at the same time as or slightly earlier than the guest; if the elevator arrives too early, resulting in a long wait, the call command can be delayed appropriately; if the elevator arrives too late, the guest will still have to wait, leading to poor scheduling efficiency. Therefore, by matching the two time values, the optimal call time can be determined, achieving synchronized arrival of the elevator and the guest, eliminating guest waiting time.

[0034] This invention constructs a basic profile and a real-time iterative profile of the target guest, and monitors the time-series behavior data of guest room doors in real time. By combining information such as guest door opening and closing actions, reservation service information, and behavioral preferences, the system proactively distinguishes the guest's stay status and travel scenario, and then calculates the confidence level of their travel intention. When the confidence level of travel intention reaches a certain threshold, the system estimates the predicted elevator arrival time based on the real-time elevator operating status and determines the timing of issuing the elevator call command based on the movement characteristics in the real-time iterative profile, ensuring the elevator arrives at the guest's floor ahead of schedule. This method transforms elevator scheduling from a passive response to proactive pre-scheduling, initiating the scheduling process before the guest leaves the room, achieving synchronous arrival of the elevator and the guest, and significantly reducing the guest's waiting time in the elevator lobby. Simultaneously, because the scheduling decision is based on the profile and door behavior, it effectively filters out false triggers, avoiding the waste of elevator resources due to ineffective scheduling, thereby improving the guest experience and optimizing elevator operating efficiency.

[0035] In one possible implementation, guest attribute data includes room floor, room number, distance from the room to the elevator lobby, historical check-in behavior, current check-out time, and the target floor corresponding to the reserved service. Based on the guest attribute data of the target guest, a basic profile and a real-time iterative profile of the target guest are constructed, including: The basic profile of the target guest is obtained by combining the target guest's room floor, room number, distance from the room to the elevator lobby, historical check-in behavior, current check-out time, and the target floor corresponding to the reservation service. The travel time points are extracted from the historical room door time sequence behavior data of the target guests, and cluster analysis is performed to obtain the high-frequency travel time periods of the target guests; Based on the statistical frequency distribution of the target guests' historical travel types, the travel type preferences of the target guests are obtained; The target floor preference of the target guests is obtained by statistically analyzing the frequency distribution of their historical travel target floor records; Extract the ratio of displacement to time from the historical movement trajectory points of the target guest, and calculate the average value to obtain the walking speed of the target guest; The probability of trip cancellation for the target guest is obtained by calculating the sliding window ratio from historical travel attempt and cancellation records; By combining high-frequency travel times, travel type preferences, target floor preferences, walking speed, and travel cancellation probability, a real-time iterative profile of the target guest is obtained.

[0036] In this embodiment, guest attribute data includes at least the guest room floor, room number, distance from the guest room to the elevator lobby, historical check-in behavior, current check-out time, and the target floor corresponding to the reservation service. These data come from various sources: the guest room floor and room number are assigned at check-in; the distance from the guest room to the elevator lobby can be pre-measured using the building floor plan and stored in the database; historical check-in behavior records the number of times the guest has stayed at this hotel or chain hotel, the average length of stay, etc.; the current check-out time is the known reservation time; and the target floor corresponding to the reservation service comes from the restaurant floor, meeting room floor, etc., booked by the guest through the app or at the front desk.

[0037] The construction of real-time iterative profiles relies primarily on statistical mining of historical behavioral data. Specifically, the system extracts the time point of each trip from the guest's historical room door event timestamps, then performs cluster analysis on these time points. For example, the K-means algorithm is used to cluster the time points into several clusters, each corresponding to a high-frequency travel time period, such as 9:00-10:00 AM or 3:00-4:00 PM. Travel type preference is obtained by statistically analyzing the proportion of each type (check-out, dining, meeting, temporary outing) in historical trips; the type with the highest proportion is considered the preference. Target floor preference is similar; the frequency of guests visiting each floor in historical trips is analyzed, and the floor with the highest frequency is the preferred floor. Walking speed is obtained using positioning devices in public areas, such as Bluetooth beacons or access control card records. When a guest moves from their room to the elevator lobby, the system records the displacement and time taken, calculates the ratio, and uses a moving average method to smooth short-term fluctuations; for example, the average speed of the last 5 trips is used as the current value. Trip cancellation probability is a sliding window proportion, statistically analyzing the percentage of trip attempts within a past period (e.g., the last 10 trips) that did not actually involve taking the elevator. Combining these statistical results creates a real-time iterative profile that dynamically reflects the current behavioral habits of guests.

[0038] Basic profiles and real-time iterative profiles can be stored in the same database or separately. They are independent of each other but work together to determine travel intent.

[0039] In one possible implementation, the guest room door time-series behavior data includes door opening and closing events; real-time monitoring of the target guest's guest room door time-series behavior data distinguishes between the target guest's stay status and travel scenario, including: Mark the first time a target guest swipes their card to enter the room after check-in as the check-in entry event and use it as the starting point of the stay period; During the stay period, if a door opening event is detected and no further door opening events occur after a preset silent window period, the door opening event is determined to be an outgoing door opening, and the stay is switched to non-stay status. If the current time is less than the time of this check-out, and the guest room door timing behavior data includes the door opening mode within the second preset threshold time after the door is closed, then it is determined to be a check-out travel scenario. If the current time is less than the time of any scheduled service and an opening event occurs in the guest room door time sequence behavior data, it is determined to be a fixed service travel scenario. If a door opening event occurs in the guest room door sequence behavior data, but does not meet the criteria for check-out or fixed service travel scenarios, it is determined to be a temporary travel scenario.

[0040] In this embodiment, the guest room door timing behavior data includes door opening events and door closing events, each with a precise timestamp.

[0041] To monitor this data in real time and differentiate between stay status and travel scenarios, it's crucial to clarify two fundamental concepts: check-in / stay period and non-stay status. The check-in / stay period refers to the time between a guest's initial entry into the room and their confirmed departure. The non-stay status refers to a state where the guest has left the room, the door is closed, and they are unlikely to return in the near future. The specific decision-making process is as follows: When the system detects a guest's first entry after check-in, it marks this event as a check-in entry event, establishing the starting point of the check-in / stay period. Subsequently, the system continuously monitors door status changes. Within the check-in / stay period, if a door opening event is detected, the system does not immediately classify it as a departure but initiates a silent window. The duration of this window can be set according to the hotel's actual environment, typically between 20 and 60 seconds. If no further door opening events occur within the silent window, and each previous opening event is inevitably followed by a closing event (for security reasons, guests always close the door after opening it), the system infers that the guest has left the room, classifying the opening event as a departure and transitioning to the non-stay status. Conversely, if another door-opening event occurs within the silent window, it indicates that the guest may have only opened the door tentatively or briefly before returning, and in this case, the resident status remains unchanged. From the moment a door-opening event occurs until the next door-opening event occurs, the system classifies the guest as being in a non-resident state.

[0042] The distinction between travel scenarios relies on the relationship between the current time and the booking event, as well as door behavior patterns. Determining a check-out travel scenario requires simultaneously meeting the following conditions: the current time is less than a first preset threshold (e.g., 30 minutes) from the booked check-out time; and the room door status change sequence shows a pattern of the door being closed and then reopened within a second preset threshold time. This pattern typically reflects the guest's behavior of packing their luggage, locking the door, leaving, and then returning to check, which is a strong signal of check-out travel. Optionally, the guest's bill settlement status (obtainable from the hotel management system) can also be used as an auxiliary factor in the determination.

[0043] Determining a fixed service travel scenario is relatively simple: if the current time is less than a third preset threshold (e.g., 15 minutes) from any scheduled service time (e.g., breakfast at 8 am or meeting at 9 am), and an opening event occurs in the room door status change, the system assumes that the guest may be going to enjoy the scheduled service.

[0044] Temporary travel scenarios serve as a fallback category. If a door opens but the current time is neither close to check-out nor has a reservation been made, and the guest is currently within their stay period, it is classified as temporary travel. This three-tiered classification method ensures rapid response in high-confidence scenarios while also preserving the possibility of scheduling for travel without pre-set information.

[0045] In one possible implementation, the confidence level of travel intention is calculated based on travel scenarios and real-time iterative profiles, including: If the scenario is determined to be a check-out travel scenario, the travel confidence score will be assigned the first preset score. If the scenario is determined to be a fixed-service travel scenario, the travel confidence score is assigned a second preset score; wherein the second preset score is less than the first preset score. If the scenario is determined to be a temporary travel scenario, the travel confidence score is obtained by weighted summation of the confidence score influencing factors. Among them, the confidence score influencing factors include at least one of the following: room door opening and closing behavior factor, historical travel frequency factor, and travel cancellation penalty factor.

[0046] In this embodiment, the calculation of travel intention confidence requires a differentiated strategy based on different travel scenarios. For check-out travel scenarios, due to their highest certainty and urgency, the system directly assigns a first preset score to the travel intention confidence, such as 90 points or above (out of 100). The reason for this approach is that the triggering conditions for check-out travel scenarios already include multiple strong signals (approaching check-out, bills settled, and opening the door immediately after closing), making misjudgment almost impossible. Giving a high score ensures that scheduling is executed immediately.

[0047] For fixed-service travel scenarios, the system assigns a second preset score to the confidence level of travel intention. This score is lower than the first preset score, for example, between 75 and 89 points. This score range reflects the system's high confidence in the booked service while also retaining a certain degree of redundancy, as guests may cancel their reservations or change their plans at the last minute.

[0048] For impromptu travel scenarios, due to the lack of strong prompts such as check-out or reservations, a weighted scoring model is needed to calculate confidence levels. This model considers multiple factors: Guest room door opening / closing behavior: if the door is opened briefly after closing (e.g., within 10 seconds), a higher score is assigned, as this often indicates a clear intention to leave; if it's simply opening the door (after it has been closed for a longer period), a base score is assigned. Historical travel frequency: if the current time period falls within the guest's high-frequency travel period, the score is calculated by multiplying the historical travel probability within that period by the corresponding weight. Trip cancellation deduction factor: if the guest frequently cancels trips historically (i.e., does not actually use the elevator after triggering pre-scheduling), a certain percentage of the score is deducted from the total score. The weights of these three factors can be set empirically, for example, 40% for the guest room door opening / closing behavior factor, 30% for the historical travel frequency factor, and 30% for the trip cancellation deduction factor. The weighted sum of the scores for each factor yields the confidence level of travel intention in impromptu travel scenarios.

[0049] Regardless of the scenario, the final confidence level will be compared with a preset confidence threshold, which is usually set to 60 points in engineering implementation.

[0050] In one possible implementation, the feature is that, after issuing the elevator call command, it further includes: If a target resident is detected entering the elevator within the fourth preset threshold time after the elevator arrives, and the elevator floor change matches the current travel scenario, then it is determined to be a valid dispatch. The walking speed in the real-time iterative profile is updated using an exponentially weighted moving average method. And / or, use event counting to update the high-frequency travel time periods and target floor preferences in the real-time iterative profile.

[0051] In this embodiment, after issuing the elevator call command, the system also needs to perform closed-loop verification to confirm the validity of the dispatch and update the guest profile accordingly. Specifically, the system continuously monitors elevator operation feedback and the subsequent status of the guest room door. If, within a fourth preset threshold timeframe after the elevator reaches the guest's floor (e.g., 30 seconds), the system detects that the guest has entered the elevator car, and the subsequent floor changes registered by the elevator match the previously determined travel scenario (e.g., the elevator eventually reaches the lobby floor in a check-out scenario, or the restaurant floor in a dining scenario), then the dispatch is considered valid, and the door opening timestamp of this trip is recorded as the travel time. Valid dispatch means that the system's prediction is correct, and the guest's profile parameters should align with the observed actual behavior.

[0052] For walking speed, the system uses an exponentially weighted moving average method for updating. This method can take into account both historical data and new observations. The calculation formula is: new speed = α × current measured speed + (1-α) × old speed, where α is usually between 0.2 and 0.3, so that the impact of new observations on the image is gradual and not drastic.

[0053] For high-frequency travel periods and target floor preferences, the system uses an event counting method for updates: Each time a valid trip occurs, the system records the travel time and target floor, then recalculates the time and floor distribution over a past period (e.g., the last 7 days or the last 10 trips), using the most frequent time periods and floors as new preference values. Through this gradual update, the real-time iterative profile can progressively adapt to changes in guest behavior; for example, a guest's walking speed may increase due to familiarity with the environment, or a temporary increase in preference for a particular target floor may occur due to attending a meeting. Continuous iteration of the profile is crucial for the method's long-term stable operation.

[0054] In one possible implementation, the feature is that, after issuing the elevator call command, it further includes: If no target guest is detected entering the elevator within the fourth preset threshold time after the elevator arrives, and the room door status shows that the door is closed again, it is determined to be an invalid dispatch. Cancel the issued elevator call command, reduce the preset confidence threshold by a preset percentage, and increase the probability of trip cancellation in the real-time iterative profile by a preset increment.

[0055] In this embodiment, if closed-loop verification detects invalid scheduling, the system needs to quickly correct and adjust its strategy. The criteria for invalid scheduling are: within a fourth preset threshold time after the elevator arrives, the system does not detect the guest entering the elevator, and the guest room door status shows it is closed again. This usually means that the guest canceled their trip for some reason after opening the door, such as returning to their room after answering a phone call, or simply opening the door to check the corridor and then closing it again. Invalid scheduling not only wastes elevator resources but may also interfere with other guests' normal elevator use. Therefore, once the system determines invalid scheduling, it should immediately send an instruction to the elevator control system to cancel the previously issued elevator call instruction, remove the elevator's stop request from the task queue, and release elevator resources for other requests.

[0056] Simultaneously, the system needs to adjust the parameters related to this guest to reduce the probability of similar false triggers in the future. Specific adjustments include: lowering the guest's confidence threshold by a preset percentage, such as 5% to 10%. This means the criteria for judging the guest's future travel will be slightly relaxed, as the system may currently be too strict and frequently misjudge. Additionally, the probability of trip cancellation in the real-time iterative profile will be increased by a preset increment, such as 0.05 or 5 percentage points. With the increased probability of trip cancellation, this factor will deduct more points in the weighted scoring model for subsequent temporary travel scenarios, thereby suppressing the system's tendency to respond to the guest's temporary travel and reducing ineffective scheduling. This two-way adjustment mechanism allows the system to adaptively change for different guest behavior styles: for guests who frequently cancel trips, the system will lower the sensitivity and confidence threshold; for guests who rarely cancel, the system will maintain a higher response level.

[0057] In one possible implementation, the method further includes: When there are multiple target guests and the current time is the peak check-out period or the peak time for meeting end, multiple elevator call requests on the same floor will be combined and processed, and elevator capacity will be allocated according to the principle of prioritizing check-out guests over regular guests and VIP members over regular members.

[0058] In this embodiment, when multiple guests simultaneously trigger pre-scheduling, and the current time period falls within the peak check-out or conference end-of-term period, the system needs to implement a multi-elevator cluster scheduling strategy to avoid elevators frequently and inefficiently shuttling between floors, thereby improving overall capacity. The peak check-out period typically occurs between 11:00 AM and 12:00 PM, when a large number of guests prepare to check out; the peak conference end-of-term period corresponds to the end time of the conference, such as 5:00 PM or 6:00 PM. A common characteristic of these peak periods is the rapid influx of guests from different floors to the lobby or parking lot.

[0059] The system first obtains the current floor and target floor information of all guests whose pre-scheduling has been triggered within the specified time period. Since most guests' target floors are the lobby or parking floors, the elevator call commands from multiple guests on the same floor (e.g., the 10th floor) can be merged into a single command, instructing the elevator to stop at the 10th floor once, instead of stopping individually for each guest. After merging, the system needs to coordinate the task allocation of multiple elevators.

[0060] An effective strategy is to allocate elevator resources by priority: guests checking out have the highest priority, followed by VIP members (such as gold and diamond card members), and finally regular guests. The system can allocate elevator resources based on these priorities; for example, one elevator can be designated specifically for guests checking out on higher floors, another for guests checking out on lower floors, and a third reserved for occasional non-check-out requests. Simultaneously, dynamic programming or greedy algorithms can be used in the scheduling algorithm to minimize elevator idle runs and reverse trips. An exemplary strategy is to sort all pending elevator calls by floor from highest to lowest, assign one elevator to higher floors and another to lower floors, while prioritizing check-out requests. In practice, a greedy algorithm can be used, selecting the elevator capable of handling the most pending requests each time.

[0061] By merging requests for the same floor and prioritizing them, the system can maximize elevator capacity during peak hours, reduce average waiting time for guests, and avoid elevator paralysis caused by a large number of simultaneous calls. This strategy works in conjunction with the aforementioned individual pre-scheduling method, employing individualized precise scheduling during normal times and switching to cluster optimization mode during peak hours, thus balancing service experience and operational efficiency.

[0062] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0063] The following are device embodiments of the present invention. For details not described in detail, please refer to the corresponding method embodiments described above.

[0064] Figure 2 A schematic diagram of a hotel elevator pre-scheduling device based on user profiles provided in an embodiment of the present invention is shown. For ease of explanation, only the parts relevant to the embodiment of the present invention are shown, and are described in detail below: like Figure 2 As shown, the hotel elevator pre-scheduling device 2 based on user profiles includes: The profile building module 21 is used to build a basic profile and a real-time iterative profile of the target guest based on the guest attribute data of the target guest; The travel judgment module 22 is used to monitor the time-series behavior data of the target guest room door in real time, distinguish the stay status and travel scenario of the target guest room door based on the time-series behavior data and basic profile, and calculate the confidence level of travel intention based on the travel scenario and real-time iterative profile. The time prediction module 23 is used to calculate the predicted time for the elevator to reach the floor where the target guest is located based on the real-time operating status of the elevator when the confidence level of the travel intention is greater than the preset confidence level threshold. The instruction issuing module 24 is used to determine the timing of issuing the elevator call instruction based on the matching relationship between walking speed and predicted time in the real-time iterative profile, so that the elevator can arrive at the floor where the target resident is located in advance.

[0065] In one possible implementation, guest attribute data includes guest room floor, guest room number, distance from guest room to elevator lobby, historical check-in behavior, current check-out time, and the target floor corresponding to the reserved service; the profile building module 21 is specifically used for: The basic profile of the target guest is obtained by combining the target guest's room floor, room number, distance from the room to the elevator lobby, historical check-in behavior, current check-out time, and the target floor corresponding to the reservation service. The travel time points are extracted from the historical room door time sequence behavior data of the target guests, and cluster analysis is performed to obtain the high-frequency travel time periods of the target guests; Based on the statistical frequency distribution of the target guests' historical travel types, the travel type preferences of the target guests are obtained; The target floor preference of the target guests is obtained by statistically analyzing the frequency distribution of their historical travel target floor records; Extract the ratio of displacement to time from the historical movement trajectory points of the target guest, and calculate the average value to obtain the walking speed of the target guest; The probability of trip cancellation for the target guest is obtained by calculating the sliding window ratio from historical travel attempt and cancellation records; By combining high-frequency travel times, travel type preferences, target floor preferences, walking speed, and travel cancellation probability, a real-time iterative profile of the target guest is obtained.

[0066] In one possible implementation, the guest room door timing behavior data includes door opening events and door closing events; the travel determination module 22 is specifically used for: Mark the first time a target guest swipes their card to enter the room after check-in as the check-in entry event and use it as the starting point of the stay period; During the stay period, if a door opening event is detected and no further door opening events occur after a preset silent window period, the door opening event is determined to be an outgoing door opening, and the stay is switched to non-stay status. If the current time is less than the time of this check-out, and the guest room door timing behavior data includes the door opening mode within the second preset threshold time after the door is closed, then it is determined to be a check-out travel scenario. If the current time is less than the time of any scheduled service and an opening event occurs in the guest room door time sequence behavior data, it is determined to be a fixed service travel scenario. If a door opening event occurs in the guest room door sequence behavior data, but does not meet the criteria for check-out or fixed service travel scenarios, it is determined to be a temporary travel scenario.

[0067] In one possible implementation, the travel determination module 22 is specifically used for: If the scenario is determined to be a check-out travel scenario, the travel confidence score will be assigned the first preset score. If the scenario is determined to be a fixed-service travel scenario, the travel confidence score is assigned a second preset score; wherein the second preset score is less than the first preset score. If the scenario is determined to be a temporary travel scenario, the travel confidence score is obtained by weighted summation of the confidence score influencing factors. Among them, the confidence score influencing factors include at least one of the following: room door opening and closing behavior factor, historical travel frequency factor, and travel cancellation penalty factor.

[0068] In one possible implementation, the image construction module 21 is further configured to: After issuing the elevator call command, if the target resident is detected to enter the elevator within the fourth preset threshold time after the elevator arrives, and the elevator floor change matches the current travel scenario, it is determined to be a valid dispatch. The walking speed in the real-time iterative profile is updated using an exponentially weighted moving average method. And / or, use event counting to update the high-frequency travel time periods and target floor preferences in the real-time iterative profile.

[0069] In one possible implementation, the image construction module 21 is further configured to: If, after issuing the elevator call command, no target guest is detected entering the elevator within the fourth preset threshold time after the elevator arrives, and the guest room door status shows that the door is closed again, then the dispatch is deemed invalid. Cancel the issued elevator call command, reduce the preset confidence threshold by a preset percentage, and increase the probability of trip cancellation in the real-time iterative profile by a preset increment.

[0070] In one possible implementation, the instruction issuing module 24 is also used for: When there are multiple target guests and the current time is the peak check-out period or the peak time for meeting end, multiple elevator call requests on the same floor will be combined and processed, and elevator capacity will be allocated according to the principle of prioritizing check-out guests over regular guests and VIP members over regular members.

[0071] This invention constructs a basic profile and a real-time iterative profile of the target guest, and monitors the time-series behavior data of guest room doors in real time. By combining information such as guest door opening and closing actions, reservation service information, and behavioral preferences, the system proactively distinguishes the guest's stay status and travel scenario, and then calculates the confidence level of their travel intention. When the confidence level of travel intention reaches a certain threshold, the system estimates the predicted elevator arrival time based on the real-time elevator operating status and determines the timing of issuing the elevator call command based on the movement characteristics in the real-time iterative profile, ensuring the elevator arrives at the guest's floor ahead of schedule. This method transforms elevator scheduling from a passive response to proactive pre-scheduling, initiating the scheduling process before the guest leaves the room, achieving synchronous arrival of the elevator and the guest, and significantly reducing the guest's waiting time in the elevator lobby. Simultaneously, because the scheduling decision is based on the profile and door behavior, it effectively filters out false triggers, avoiding the waste of elevator resources due to ineffective scheduling, thereby improving the guest experience and optimizing elevator operating efficiency.

[0072] Figure 3 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. For example... Figure 3 As shown, the electronic device 3 of this embodiment includes a processor 30 and a memory 31. The memory 31 stores a computer program 32. When the processor 30 executes the computer program 32, it implements the steps in the various method embodiments described above. Alternatively, when the processor 30 executes the computer program 32, it implements the functions of each module / unit in the various device embodiments described above.

[0073] For example, computer program 32 may be divided into one or more modules / units, which are stored in memory 31 and executed by processor 30 to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of computer program 32 in electronic device 3.

[0074] Electronic device 3 may include, but is not limited to, processor 30 and memory 31. Those skilled in the art will understand that... Figure 3 This is merely an example of electronic device 3 and does not constitute a limitation on electronic device 3. It may include more or fewer components than shown, or combine certain components, or different components. For example, electronic device 3 may also include input / output devices, network access devices, buses, etc.

[0075] The processor 30 can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0076] The memory 31 can be an internal storage unit of the electronic device 3, such as a hard disk or memory of the electronic device 3. The memory 31 can also be an external storage device of the electronic device 3, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the electronic device 3. Furthermore, the memory 31 can include both internal and external storage units of the electronic device 3. The memory 31 is used to store the computer program 32 and other programs and data required by the electronic device 3. The memory 31 can also be used to temporarily store data that has been output or will be output.

[0077] For the sake of simplicity and clarity, only the above-described functional modules / units are used as examples. In practical applications, the functions described above can be assigned to different functional modules / units as needed. These modules / units can be implemented in hardware, software, or a combination of both.

[0078] This invention also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the methods described in the above-described method embodiments.

[0079] This invention also provides a computer program product, including a computer program. When the computer program is executed by a processor, it implements the methods described in the above-described method embodiments.

[0080] Computer programs include computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. Computer-readable media can include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.

[0081] In the above embodiments, the descriptions of each embodiment have their own emphasis. Parts not detailed or described in a particular embodiment can be referred to in the relevant descriptions of other embodiments. Unless otherwise specified or in conflict with logic, the terminology and / or descriptions between different embodiments are consistent and can be referenced interchangeably. Technical features in different embodiments can be combined to form new embodiments based on their inherent logical relationships.

[0082] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A hotel elevator pre-scheduling method based on user profiles, characterized in that, include: The basic profile and real-time iterative profile of the target guest are constructed based on the guest attribute data of the target guest; Real-time monitoring of the target guest’s door-time behavior data, distinguishing the target guest’s stay status and travel scenario based on the door-time behavior data and the basic profile, and calculating the travel intention confidence level based on the travel scenario and the real-time iterative profile; If the confidence level of the travel intention is greater than the preset confidence level threshold, the predicted time for the elevator to reach the floor where the target guest is located is calculated based on the real-time operating status of the elevator. Based on the matching relationship between walking speed in the real-time iterative profile and the predicted time, the timing of issuing the elevator call command is determined so that the elevator arrives at the floor where the target resident is located in advance.

2. The hotel elevator pre-scheduling method based on user profiles according to claim 1, characterized in that, The guest attribute data includes the room floor, room number, distance from the room to the elevator lobby, historical check-in behavior, current check-out time, and the target floor corresponding to the reservation service; The construction of the basic profile and real-time iterative profile of the target guest based on the guest attribute data of the target guest includes: The basic profile of the target guest is obtained by combining the target guest's room floor, room number, distance from the room to the elevator lobby, historical check-in behavior, current check-out time, and the target floor corresponding to the reservation service. The travel time points are extracted from the historical room door time sequence behavior data of the target guests, and cluster analysis is performed to obtain the high-frequency travel time periods of the target guests; Based on the statistical frequency distribution of the target guest's historical travel types, the target guest's travel type preference is obtained; The target floor preference of the target guests is obtained by statistically analyzing the frequency distribution from their historical travel target floor records; The ratio of displacement to time is extracted from the historical travel trajectory points of the target guest, and the average value is calculated to obtain the walking speed of the target guest; The probability of trip cancellation for the target guest is obtained by calculating the sliding window ratio from historical travel attempt and cancellation records; The real-time iterative profile of the target guest is obtained by combining the high-frequency travel time period, the travel type preference, the target floor preference, the walking speed, and the travel cancellation probability.

3. The hotel elevator pre-scheduling method based on user profiles according to claim 1, characterized in that, The guest room door timing behavior data includes door opening events and door closing events; The real-time monitoring of the target guest's door-time behavior data to distinguish the target guest's stay status and travel scenario includes: The door opening event when the target guest first swipes the door to enter after check-in is marked as the check-in entry event and is used as the starting point of the stay period; During the stay period, if a door opening event is detected and no further door opening events occur after a preset silent window period, the door opening event is determined to be an exit door opening, and the stay is switched to a non-stay state. If the current time is less than the time of this check-out, and the guest room door timing behavior data includes the door opening mode within a second preset threshold time after the door is closed, then it is determined to be a check-out travel scenario. If the current time is less than the time of any scheduled service and an opening event occurs in the guest room door timing behavior data, it is determined to be a fixed service travel scenario. If an opening event occurs in the guest room door timing behavior data, and the conditions for determining a check-out travel scenario or a fixed service travel scenario are not met, then it is determined to be a temporary travel scenario.

4. The hotel elevator pre-scheduling method based on user profiles according to claim 3, characterized in that, The calculation of travel intention confidence based on the travel scenario and the real-time iterative profile includes: If the scenario is determined to be a check-out travel scenario, the travel confidence score will be assigned the first preset score. If the scenario is determined to be a fixed-service travel scenario, the travel confidence score is assigned a second preset score; wherein the second preset score is less than the first preset score. If the scenario is determined to be a temporary travel scenario, the travel confidence score is obtained by weighted summation of the confidence score influencing factors; wherein, the confidence score influencing factors include at least one of the following: guest room door opening and closing behavior factor, historical travel frequency factor, and travel cancellation penalty factor.

5. The hotel elevator pre-scheduling method based on user profiles according to any one of claims 1 to 4, characterized in that, After issuing the elevator call command, the following is also included: If the target guest is detected to enter the elevator within a fourth preset threshold time after the elevator arrives, and the elevator floor change matches the current travel scenario, then it is determined to be a valid dispatch. The walking speed in the real-time iterative profile is updated using an exponentially weighted moving average method. And / or, use event counting to update the high-frequency travel time periods and target floor preferences in the real-time iterative profile.

6. The hotel elevator pre-scheduling method based on user profiles according to any one of claims 1 to 4, characterized in that, After issuing the elevator call command, the following is also included: If the target guest is not detected to enter the elevator within the fourth preset threshold time after the elevator arrives, and the room door status shows that the door is closed again, it is determined to be an invalid dispatch. The issued elevator call command is revoked, the preset confidence threshold is reduced by a preset percentage, and the probability of trip cancellation in the real-time iterative profile is increased by a preset increment.

7. The hotel elevator pre-scheduling method based on user profiles according to any one of claims 1 to 4, characterized in that, The method further includes: When there are multiple target guests and the current time is the peak check-out period or the peak time for meeting end, multiple elevator call instructions on the same floor will be combined and processed, and elevator capacity will be allocated according to the principle of prioritizing check-out guests over regular guests and VIP members over regular members.

8. A hotel elevator pre-scheduling device based on user profiles, characterized in that, include: The profile building module is used to build a basic profile and a real-time iterative profile of the target guest based on the guest attribute data of the target guest; The travel judgment module is used to monitor the time-series behavior data of the target guest's room door in real time, distinguish the stay status and travel scenario of the target guest based on the time-series behavior data of the room door and the basic profile, and calculate the travel intention confidence based on the travel scenario and the real-time iterative profile. The time prediction module is used to calculate the predicted time for the elevator to reach the floor where the target guest is located based on the real-time operating status of the elevator when the confidence level of the travel intention is greater than a preset confidence level threshold. The instruction issuance module is used to determine the timing of issuing an elevator call instruction based on the matching relationship between the walking speed in the real-time iterative profile and the predicted time, so that the elevator can arrive at the floor where the target resident is located in advance.

9. An electronic device, characterized in that, It includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method as described in any one of claims 1 to 7.

Images