Intelligent restaurant ordering system based on internet of things

Abstract

This invention relates to the fields of Internet of Things (IoT) and data processing technology, and discloses an IoT-based smart restaurant ordering system. It aims to solve the problems of system concurrency bottlenecks and excessive ineffective network communication overhead caused by the lag in physical status information between the front-end and back-end. The system utilizes IoT modules to collect real-time physical operating status parameters of kitchen processing equipment and the real-time weight of food containers. The back-end server uses this data to build a model, calculate the expected processing latency and the slope of food consumption rate, and generate dynamic rendering instructions. The ordering terminal receives these instructions and adjusts the rendering level, interactive availability status, and visual warning labels of the corresponding dishes in real time. This invention achieves direct feedback control of the underlying physical data to the front-end UI presentation layer, effectively reducing the system's database retrieval pressure.

Description

Technical Field

[0001] This invention relates to the fields of Internet of Things (IoT) and data processing technology, and discloses an IoT-based smart restaurant ordering system. Background Technology

[0002] With the development of information technology in the catering industry, QR code ordering systems have been widely adopted. However, most existing smart restaurant ordering systems adopt an interactive architecture based on static databases, and the display status of the front-end menu relies heavily on manual back-end data entry or simple inventory deductions based on digital ledgers. This information flow mechanism, detached from the physical environment, suffers from serious timeliness defects. The ordering front end cannot perceive the real physical load conditions of the kitchen processing equipment or the instantaneous physical consumption rate of ingredients in containers, resulting in a long-term disconnect between the usability display of the front-end digital interface and the actual physical processing capacity at the kitchen level.

[0003] During peak business periods, the aforementioned architectural flaws directly lead to system data processing congestion and wasted underlying computing resources under high concurrency environments. Because the system lacks the ability to dynamically feed forward and intervene in front-end order placement based on the backend's physical state, a large number of user terminals continuously send invalid concurrent order requests to processing nodes that are already at full capacity or to ingredient nodes that are nearing depletion. To passively compensate for this information gap, existing system terminals are typically forced to adopt high-frequency full interface refresh or polling strategies. This not only causes an exponential increase in server network communication overhead and database concurrent read / write pressure but also results in a large amount of redundant data packet transmission.

[0004] In summary, the core technological flaw of existing mainstream smart ordering systems lies in their limitation to mere software code representing business transaction rules, lacking a closed-loop feedback control link between the IoT sensing layer and the front-end application presentation layer. This system architecture, based on unidirectional and lagging data flow, inevitably leads to significant waste of network communication bandwidth and database retrieval computing power, and it remains unable to fundamentally overcome the severe imbalance between high-concurrency data requests from multiple terminals and the limited physical execution resources of restaurants. Summary of the Invention

[0005] The purpose of this invention is to provide a smart restaurant ordering system based on the Internet of Things to solve the problems in the background art.

[0006] The objective of this invention can be achieved through the following technical solutions: A smart restaurant ordering system based on the Internet of Things (IoT) includes: The Internet of Things (IoT) sensing module is used to collect real-time physical operating status parameters of processing equipment in the kitchen and real-time weight data of food containers in the food preparation station. A backend processing server is communicatively connected to the IoT sensing module. The backend processing server is configured to: construct a device load assessment model based on the physical operating status parameters of the processing equipment, and calculate the expected processing delay for each dish; calculate the current available inventory and consumption rate slope of the ingredients associated with each dish based on the real-time weight data and a preset dish consumption model; and generate dynamic rendering instructions for the front-end ordering interface based on the expected processing delay and the consumption rate slope using a multi-criteria decision algorithm. The user ordering terminal is used to receive the dynamic rendering instructions via a wireless network and adjust the rendering level, interactive availability status, or visual warning labels of each dish in the display interface in real time according to the dynamic rendering instructions.

[0007] Preferably, the physical operating status parameters of the processing equipment collected by the IoT sensing module include: the operating current characteristics of the processing equipment, temperature probe data, and the continuous operating time of the current work cycle; The IoT sensing module collects real-time weight data through a gravity sensor array deployed at the bottom of the food container, and packages and uploads the collected data sequence to the backend processing server according to an adaptive time period.

[0008] Preferably, the device load assessment model constructed by the backend processing server specifically includes: Extract the current temperature probe data and operating current characteristics of each processing device to determine the current physical operation stage of the processing device, wherein the physical operation stage includes preheating, full-load processing, or idle stage; Query the current queue of pending orders assigned to the target processing equipment; The physical operation phase is converted into a basic processing time base, and combined with the length of the queued order and the historical average processing cycle of the target processing equipment, the expected processing delay corresponding to each dish if an order is generated at the current moment is calculated.

[0009] Preferably, the specific steps for the backend processing server to calculate the current available inventory and consumption rate slope of the ingredients associated with each dish include: Construct a sliding data window based on time series; Extract multiple consecutive real-time weight data of the food container within the sliding data window, calculate the weight drop difference between adjacent time points, and obtain the current available inventory. The weight drop difference was linearly fitted to the corresponding time span using a regression analysis algorithm, and the absolute value of the slope of the fitted trajectory was extracted as the slope of the consumption rate of the corresponding food ingredient.

[0010] Preferably, the specific execution logic of the multi-criteria decision algorithm includes: The calculated expected processing delay is compared with a preset first delay threshold, and the consumption rate slope is compared with a preset second slope threshold. When the expected processing delay exceeds the first delay threshold, or the consumption rate slope exceeds the second slope threshold and the current available inventory is lower than the preset safety stock level, it is determined that the corresponding dish triggers a high-pressure flow restriction state. Based on the extent to which the expected processing delay exceeds the first delay threshold, or the extent to which the consumption rate slope exceeds the second slope threshold, determine the current limiting level corresponding to the high-pressure current limiting state, and assign a display weight degradation coefficient that matches the current limiting level to the dish; The control logic containing the display weight degradation coefficient is encapsulated into the dynamic rendering instruction.

[0011] Preferably, the user ordering terminal adjusts the rendering level of each dish in the display interface in real time according to the dynamic rendering instruction, specifically including: Analyze the display weight degradation coefficient carried in the dynamic rendering instruction; Dishes with higher display weights will be sorted in a grid or list view on the front-end interface in a dimensionality reduction manner. The dimensionality reduction sorting process includes: forcibly moving the UI component of the corresponding dish to the end of the display list, or collapsing and hiding it in a sub-menu node that requires a second click to expand.

[0012] Preferably, the user ordering terminal adjusts the visual warning labels of each dish on the display interface according to the dynamic rendering instruction, specifically including: When the expected processing delay calculated by the backend processing server falls into different preset risk ranges, the dynamic rendering instruction drives the user ordering terminal to overlay congestion warning watermarks with different color codes on the corresponding main image of the dish; or, above the details interaction layer of the corresponding dish, a floating text mask layer containing a dynamic countdown of the expected waiting time is rendered.

[0013] Preferably, the user ordering terminal adjusts the interactive availability status of each dish on the display interface according to the dynamic rendering instruction, specifically including: When it is determined that the current available stock of the ingredients associated with the target dish cannot meet the minimum standard serving size, or the corresponding processing equipment is in a physical malfunction and out of service; The dynamic rendering instruction triggers the user ordering terminal to perform a graying-out and freezing operation on the "Add to Cart" or "Order Now" interactive control for the target dish, and blocks the user's touch click event on the control at the local interception layer of the terminal operating system.

[0014] Preferably, the backend processing server is further configured to perform a model reverse optimization mechanism based on the actual processed data: The system log records the predicted value of the expected processing delay and the actual physical time taken from order dispatch to meal preparation for actual users. Calculate the residual between the predicted value and the actual physical time consumption data; When the average residual value within the preset analysis period exceeds the tolerance threshold, the backend processing server uses the residual value to perform reverse compensation update on the weight calculation coefficients in the device load assessment model.

[0015] Preferably, the backend processing server and the user ordering terminal use an event-driven incremental push mechanism when transmitting the dynamic rendering instructions: The backend processing server sends a data packet for a single dish to the user's ordering terminal that is currently in an active network connection state only when the expected processing delay for a certain dish exceeds the set delay level, or when the consumption rate slope changes abruptly. This triggers the terminal to perform a redraw update of local components to avoid data redundancy caused by a full menu refresh.

[0016] The beneficial effects of this invention are: This invention directly collects data on the current and temperature of kitchen processing equipment, as well as the physical weight of ingredients, through an IoT sensing module, breaking away from the reliance of traditional ordering systems on manual data entry or static databases. By using a load assessment and consumption rate model built on the backend server, the system can translate the underlying physical operating conditions into real-time rendering instructions for the front-end UI, allowing the ordering interface to dynamically adjust the weight of dishes based on the actual production pressure in the kitchen. This feedback loop from physical entities to the presentation layer ensures a high degree of physical realism in the waiting time and inventory status seen by consumers, optimizing dining expectations from the outset.

[0017] The system utilizes an event-driven incremental push mechanism to replace the high-frequency, full-scale polling strategy of traditional architectures. Data transmission is triggered only when there are critical changes in the physical state, greatly reducing invalid network packet interactions. Simultaneously, by automatically sorting, collapsing, or interactively freezing high-pressure, flow-limited dishes on the front-end interface, the system leverages a feedforward intervention mechanism at the UI layer to divert traffic before users place orders, effectively reducing the concurrent retrieval pressure on the underlying database from invalid requests and ensuring system stability during peak business periods.

[0018] This invention introduces a model back-end optimization mechanism based on residual analysis, which automatically updates the weight coefficients of the back-end evaluation model by comparing the deviation between the expected processing delay and the actual food preparation time. This closed-loop optimization capability enables the system to adapt to the impact of different chefs' work rhythms and different time periods on processing efficiency, making delay prediction increasingly accurate. Combined with visual warning labels and dynamic countdown functions, it not only enhances the user experience but also provides restaurant managers with digital scheduling data based on real-time data, achieving a deep integration of technology and business scenarios. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the structure of a smart restaurant intelligent ordering system based on the Internet of Things according to the present invention. Detailed Implementation

[0020] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0021] Example: Figure 1 As shown, this embodiment of the invention provides a smart restaurant ordering system based on the Internet of Things. The overall system architecture is divided into three layers from bottom to top: physical perception layer (Internet of Things sensing module), core computing layer (back-end processing server), and application presentation layer (user ordering terminal).

[0022] The IoT sensing module is deployed in the back kitchen of the smart restaurant and is mainly responsible for high-frequency data collection of the physical status of processing equipment and food containers in the food preparation station.

[0023] For each processing machine, the system is equipped with a non-invasive current sensor based on the Hall effect to collect the operating current characteristics. High-precision thermocouple temperature probes are used to collect the temperature inside the cavity or on the heating plate. And an embedded timer module, used to record the duration of the current job loop. .

[0024] A high-precision gravity sensor array is deployed at the bottom of a standard GN basin in the food preparation area. This array consists of multiple miniature strain gauge load cells connected in parallel and linked to an HX711 analog-to-digital converter chip. To reduce network load, the gravity sensors do not use a fixed-frequency upload frequency; instead, they employ an adaptive time-period packetization strategy: when weight changes are gradual, the upload baseline period is set to [value missing]. When a sudden change in weight is detected, such as when a chef is grabbing ingredients and the first derivative exceeds a set threshold, the sampling and packaging cycle is automatically shortened. .

[0025] After receiving the device status data uploaded by the IoT sensing module, the backend processing server calculates the expected processing delay required if a user places an order for a certain dish at the current moment by building a device load assessment model.

[0026] The server first analyzes the operating current. With temperature data Determine the current physical operating stage of the equipment The physical stages are defined as follows: preheating stage. Full-load processing stage With idle phase .

[0027] when In the rising phase and When it is in the peak range, it is determined to be or Further distinctions are made based on whether data exists in the queue.

[0028] To calculate a specific target dish Expected processing delay The system defines the following core formulas: in, Indicates the target dish Total expected delay from order generation to completion of processing. This indicates the base processing time of the target processing equipment based on the current physical operating stage. For example, if the equipment is in... (Preheating), this value represents the remaining theoretical time required to reach the set operating temperature; if in the preheating stage... (Idle), this value is the device startup wake-up time; if in If the load is full, this value represents the remaining time of the currently executing job cycle, which is the standard single job time minus the time already spent running. . This represents the length of the queue of orders that have been assigned to the target processing equipment but have not yet started processing. `i` is the index variable of the `i`th dish waiting in the queue. Let i be the i-th specific dish in the queue. Process dishes for the target processing equipment recorded in the backend database. The historical average processing cycle time. This is the penalty factor for heat decay under high load on the equipment. When the equipment operates continuously at full load for a certain period of time, its heating efficiency will decrease. It increases nonlinearly with the increase of continuous working time of the equipment. Target dish The standard processing time required by the product itself.

[0029] Through this model, the system can integrate the actual "queueing situation" in the kitchen with the "physical thermal state of the machine" to obtain a waiting time that is of great reference value.

[0030] Regarding food consumption, since restaurants are in a dynamic and concurrent state, the simple "initial inventory - quantity sold" cannot reflect the true physical inventory. This invention uses physical weighing to establish a dynamic model.

[0031] Define a sliding data window based on time series The window length is L. The server extracts the real-time weight data sequence within this window. The formula for calculating the current available inventory of ingredients associated with the target dish is as follows: ,in, This represents the effective physical availability of a certain related ingredient for making the target dish at the current moment. This is the real-time total weight reading of the sensor at the latest time point within the sliding data window. This refers to the tare weight of the corresponding food container / weight of the empty bowl, stored in the backend database. The standard weight quota of this type of ingredient required to prepare a single serving of the target dish.

[0032] To predict the risk of concurrent consumption in the near future, the system uses a linear regression analysis algorithm to calculate the absolute value of the slope of the weight drop difference: in, The slope of the current physical consumption rate of the ingredient represents the extreme intensity of its consumption per unit time. n is the total number of valid sampling time nodes within the sliding data window. j is the index of the sampling node, ranging from 1 to n. Let be the time span variable of the j-th sampling node relative to the start point of the window. To be at the time node The cumulative weight drop difference between adjacent nodes represents the cumulative consumption of ingredients. The larger the absolute value of the slope calculated by this formula, the higher the frequency and quantity of the ingredient being taken in the kitchen at the current moment, meaning the greater the pressure of concurrent ordering.

[0033] After obtaining the expected processing delay Current available inventory and the slope of the consumption rate Then, the backend server uses a multi-criteria decision algorithm to generate dynamic rendering instructions for the user's ordering terminal.

[0034] Decision-making logic and calculation of weighted downgrade coefficient: Set the preset first delay threshold as The second slope threshold is Safety stock level is .

[0035] When the system determines or simultaneously satisfy At that time, the dish triggered a "high-voltage current limiting state".

[0036] To quantify this rate-limiting state so that the front-end can perform UI downgrading, the system calculates the display weight degradation coefficient: ,in, This is the display weight reduction factor assigned to the target dish; this value is a non-negative real number. The larger the value, the greater the degree to which the dish is displayed in a reduced-dimensional manner on the front end. This is the penalty weighting coefficient for processing delay factors. It is a rectification function that ensures that only the portion exceeding the threshold will generate a degradation factor; when the threshold is not exceeded, the contribution is 0. This is the baseline for the maximum user wait time that the system can tolerate. The penalty weighting coefficient for the rate of food consumption. The critical point at which the system can tolerate the maximum rate of food consumption.

[0037] The backend server will contain the dish ID and its corresponding... The value control logic is encapsulated into dynamic rendering instructions in JSON or Protobuf format.

[0038] After receiving dynamic rendering instructions, the user ordering terminal, WeChat mini program, iOS / Android native APP, or H5 page will perform the following three UI interventions in real time through the front-end framework's reactive state management: In the parsing instructions .for The dishes should be displayed using a standard waterfall or grid layout. When this happens, the front-end performs a "soft degradation," forcibly and smoothly moving its UI components from the current visible area to the bottom of the current category list; when When this happens, a "hard downgrade" is triggered, directly collapsing and hiding the dish in a second-level submenu node called "High-load dishes". Users must perform a second click to see the dish, thereby reducing the user's order conversion rate from a visual perspective.

[0039] Congestion warning watermark: based on expected processing delay The area in question will be covered with a color-coded mask of varying opacity above the main image of the dish. For example... Overlay an orange "Hot Queues" watermark; Overlay a red watermark that reads "Severe congestion in the kitchen".

[0040] Floating countdown text overlay: Rendered directly above the interactive layer on the menu details page. The converted dynamic countdown module counts down every second, giving users the most intuitive physical expectation of waiting.

[0041] If the instruction indicates the corresponding ingredient Or, if the equipment malfunctions physically, the command forces the terminal to inject [something] into the "Add to Cart" or "Order Now" button for that dish. The attribute is visually represented in gray.

[0042] To prevent malicious requests from bypassing the front-end UI and directly calling the interface, the instruction will register the dish's blacklist in the operating system's local interception layer, completely blocking the conversion of high-frequency click events initiated by users into network request packets.

[0043] Because different chefs have varying skill levels, the actual execution efficiency in the kitchen can fluctuate. This invention's server incorporates a machine learning reverse optimization mechanism, which uses system logs to analyze predicted... Compare with the actual physical time taken for the order.

[0044] Real physical time Defined as: the actual time span from when the server sends the order dispatch instruction to KDS to when the chef or device clicks / triggers the "meal preparation completed" status.

[0045] The residual values ​​are calculated as follows: ; Based on this, the system updates the historical average processing cycle time parameter in the model through a gradient descent compensation mechanism: ,in, Let be the residual value between the predicted processing delay of the k-th order and the actual physical processing time. k is the order sample number within the preset analysis period. This is to rewrite the standard operating time of the target food processing equipment in the database after the reverse compensation update. The time spent on the old standard operating procedures before the update. The learning rate or compensation tolerance coefficient controls the step size of each update, preventing model oscillations caused by a single extreme random event. M represents the total number of valid order samples for the target dish or target equipment within the preset analysis period.

[0046] When the absolute value of the average residual value within a period exceeds the tolerance threshold, the system will automatically execute the above update formula to make the model more and more in line with the actual physical operation rhythm of the kitchen.

[0047] In terms of transmission network architecture, this invention abandons the traditional HTTP long polling or high-frequency timed fetching mechanism. A persistent bidirectional long connection based on WebSocket or Server-SentEvents is established between the backend server and all active user ordering terminals.

[0048] The system incorporates an event-driven state machine. The server does not send the full menu data to the terminal; instead, it maintains a shadow state tree based on Redis. Incremental data packets are only constructed when the following two abrupt events occur: A certain dish The latency level exceeded the system's preset limit. Minor latency fluctuations do not trigger push notifications.

[0049] The slope of the consumption rate of a certain food ingredient sudden surge or Reset to zero.

[0050] When the above conditions are met, the server, based on the current routing table, sends an extremely lightweight incremental patch to all active terminals browsing the merchant's page. Upon receiving the patch, the terminal only triggers a partial redraw of the corresponding virtual DOM node, without refreshing the entire page's menu image. This mechanism significantly saves the wireless communication bandwidth of the restaurant's AP nodes and fundamentally eliminates the redundant computing power consumed when the database handles full refresh requests.

[0051] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0052] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An Internet of Things-based smart restaurant intelligent ordering system, characterized in that, include: The Internet of Things (IoT) sensing module is used to collect real-time physical operating status parameters of processing equipment in the kitchen and real-time weight data of food containers in the food preparation station. A backend processing server is communicatively connected to the IoT sensing module. The backend processing server is configured to: construct a device load assessment model based on the physical operating status parameters of the processing equipment, and calculate the expected processing delay for each dish; calculate the current available inventory and consumption rate slope of the ingredients associated with each dish based on the real-time weight data and a preset dish consumption model; and generate dynamic rendering instructions for the front-end ordering interface based on the expected processing delay and the consumption rate slope using a multi-criteria decision algorithm. The user ordering terminal is used to receive the dynamic rendering instructions via a wireless network and adjust the rendering level, interactive availability status, or visual warning labels of each dish in the display interface in real time according to the dynamic rendering instructions.

2. The smart restaurant ordering system based on the Internet of Things according to claim 1, characterized in that, The physical operating status parameters of the processing equipment collected by the IoT sensing module include: the operating current characteristics of the processing equipment, temperature probe data, and the continuous running time of the current work cycle. The IoT sensing module collects real-time weight data through a gravity sensor array deployed at the bottom of the food container, and packages and uploads the collected data sequence to the backend processing server according to an adaptive time period. 3.The smart ordering system of a smart restaurant based on the Internet of Things according to claim 2, characterized in that, The device load assessment model constructed by the backend processing server specifically includes: Extract the current temperature probe data and operating current characteristics of each processing device to determine the current physical operation stage of the processing device, wherein the physical operation stage includes preheating, full-load processing, or idle stage; Query the current queue of pending orders assigned to the target processing equipment; The physical operation phase is converted into a basic processing time base, and combined with the length of the queued order and the historical average processing cycle of the target processing equipment, the expected processing delay corresponding to each dish if an order is generated at the current moment is calculated.

4. The smart restaurant ordering system based on the Internet of Things according to claim 1, characterized in that, The specific steps for the backend processing server to calculate the current available inventory and consumption rate slope of the ingredients associated with each dish include: Construct a sliding data window based on time series; Extract multiple consecutive real-time weight data of the food container within the sliding data window, calculate the weight drop difference between adjacent time points, and obtain the current available inventory. The weight drop difference was linearly fitted to the corresponding time span using a regression analysis algorithm, and the absolute value of the slope of the fitted trajectory was extracted as the slope of the consumption rate of the corresponding food ingredient.

5. The smart ordering system of a smart restaurant based on the Internet of Things according to claim 4, characterized in that, The specific execution logic of the multi-criteria decision-making algorithm includes: The calculated expected processing delay is compared with a preset first delay threshold, and the consumption rate slope is compared with a preset second slope threshold. When the expected processing delay exceeds the first delay threshold, or the consumption rate slope exceeds the second slope threshold and the current available inventory is lower than the preset safety stock level, it is determined that the corresponding dish triggers a high-pressure current limiting state. Based on the extent to which the expected processing delay exceeds the first delay threshold, or the extent to which the consumption rate slope exceeds the second slope threshold, determine the current limiting level corresponding to the high-pressure current limiting state, and assign a display weight degradation coefficient that matches the current limiting level to the dish; The control logic containing the display weight degradation coefficient is encapsulated into the dynamic rendering instruction.

6. The smart restaurant ordering system based on the Internet of Things according to claim 5, characterized in that, The user ordering terminal adjusts the rendering level of each dish in the display interface in real time according to the dynamic rendering instructions, specifically including: Analyze the display weight degradation coefficient carried in the dynamic rendering instruction; Dishes with higher display weights will be sorted in a grid or list view on the front-end interface in a dimensionality reduction manner. The dimensionality reduction sorting process includes: forcibly moving the UI component of the corresponding dish to the end of the display list, or collapsing and hiding it in a sub-menu node that requires a second click to expand.

7. The smart restaurant ordering system based on the Internet of Things according to claim 1, characterized in that, The user ordering terminal adjusts the visual warning labels of each dish on the display interface according to the dynamic rendering instructions, specifically including: When the expected processing delay calculated by the backend processing server falls into different preset risk ranges, the dynamic rendering instruction drives the user ordering terminal to overlay congestion warning watermarks with different color codes on the corresponding main image of the dish; or, above the details interaction layer of the corresponding dish, a floating text mask layer containing a dynamic countdown of the expected waiting time is rendered.

8. The smart restaurant ordering system based on the Internet of Things according to claim 1, characterized in that, The user ordering terminal adjusts the interactive availability status of each dish on the display interface according to the dynamic rendering instructions, specifically including: When it is determined that the current available stock of the ingredients associated with the target dish cannot meet the minimum standard serving size, or the corresponding processing equipment is in a physical malfunction and out of service; The dynamic rendering instruction triggers the user ordering terminal to perform a graying-out and freezing operation on the "Add to Cart" or "Order Now" interactive control for the target dish, and blocks the user's touch click event on the control in the local interception layer of the terminal operating system.

9. A smart restaurant ordering system based on the Internet of Things according to claim 1, characterized in that, The backend processing server is also configured to perform a model reverse optimization mechanism based on the actual processed data: The system log records the predicted value of the expected processing delay and the actual physical time taken from order dispatch to meal preparation for actual users. Calculate the residual between the predicted value and the actual physical time consumption data; When the average residual value within the preset analysis period exceeds the tolerance threshold, the backend processing server uses the residual value to perform reverse compensation update on the weight calculation coefficients in the device load assessment model.

10. A smart restaurant ordering system based on the Internet of Things according to claim 1, characterized in that, The backend processing server and the user ordering terminal use an event-driven incremental push mechanism to transmit the dynamic rendering instructions: The backend processing server sends a data packet for a single dish to the user's ordering terminal that is currently in an active network connection state only when the expected processing delay for a certain dish exceeds the set delay level, or when the consumption rate slope changes abruptly. This triggers the terminal to perform a redraw update of local components to avoid data redundancy caused by a full menu refresh.

Images