Multi-station vertical conveying device and using method thereof

Abstract

The invention provides a multi-station vertical conveying device and a use method thereof, and the device comprises a system control center which is used for processing a global task instruction, distributing system resources, calculating an optimal conveying path, and generating a synchronous control instruction set; the conveying units are used for conveying materials or workpieces in the vertical direction; the working units are used for executing predetermined machining, detecting, assembling or storing operation on the materials or the workpieces conveyed by the conveying unit; the real-time sensing and communication network is used for bidirectionally transmitting whole system information including task state, real-time position and load of a conveying unit, busy and idle states of a working unit and equipment health data; according to the multi-station vertical conveying device, a centralized system control center is constructed, operation is conducted through a dynamic scheduling algorithm, unified management and control over the conveying units and all the distributed working units can be achieved, and therefore the overall operation efficiency and the response speed of the multi-station vertical conveying device are greatly improved.

Description

Technical Field

[0001] This invention relates to the field of conveying device technology, and in particular to a multi-station vertical conveying device and its usage method. Background Technology

[0002] In high-end precision manufacturing fields such as modern semiconductor manufacturing, photovoltaic cell production, and flat panel display, the clean, efficient, and non-destructive transport of wafers (such as thin and brittle sheet materials like silicon wafers and glass substrates) is one of the production processes. These industries typically adopt multi-layer cleanroom layouts, and wafers need to be frequently vertically transferred between multiple process workstations on different floors (such as cleaning, coating, photolithography, etching, and inspection).

[0003] Existing vertical conveying systems for wafer handling often employ relatively independent or predefined path control strategies, such as configuring dedicated vertical lifts for specific process lines or using fixed-cycle aerial robots. These methods suffer from poor coordination between conveying equipment and workstations along the line, lacking comprehensive real-time scheduling and optimization. When a workstation experiences status changes due to fluctuations in processing time, temporary equipment failures, or the insertion of high-priority emergency batches, the conveying system often fails to detect and dynamically adjust in real time. This can easily lead to wafer accumulation at fault points or busy workstations, increasing the risk of breakage and contamination, while also causing subsequent workstations to be idle, resulting in production line blockages or resource idleness. In wafer handling production, which is extremely sensitive to cleanliness and time, such delays and uncertainties can lead to significant economic losses. Furthermore, traditional control methods lack a global optimization perspective, typically only achieving a single "point-to-point" conveying task, making it difficult to intelligently allocate tasks and replan paths based on the real-time load of the entire system. This fails to meet the comprehensive requirements of modern intelligent manufacturing for wafer handling conveying—high flexibility, high reliability, high efficiency, and low energy consumption.

[0004] Therefore, in response to the problems mentioned above, this invention proposes a multi-station vertical conveying device and its usage method. Summary of the Invention

[0005] In order to overcome the problems of poor coordination, slow response and weak dynamic adjustment capability in the existing vertical conveying process of wafers, this invention proposes a multi-station vertical conveying device and its usage method.

[0006] The technical solution of this invention is: a multi-station vertical conveying device, comprising: The system control center is used to process global task instructions, allocate system resources, calculate the optimal transmission path, and generate a set of synchronization control instructions. Multiple conveying units are controlled by the system control center. Each conveying unit includes an independently controllable vertical lifting mechanism and its supporting platform for conveying the wafers in the vertical direction. Multiple work units, each equipped with a local controller and communicating in real time with the system control center, are used to perform predetermined processing, inspection, assembly or storage operations on the wafers delivered by the conveyor unit; A real-time sensing and communication network connects the system control center, all conveying units, and all working units, enabling bidirectional transmission of system-wide information, including task status, real-time location and load of conveying units, busy / idle status of working units, and equipment health data.

[0007] Preferably, the system control center includes a task management module, a resource status mapping module, and a path planning and scheduling module. The task management module is used to receive, parse, and queue task instructions issued by the external production management system, and assign a unique identifier and priority to each task. The resource status mapping module is used to maintain a dynamically updated digital mapping table.

[0008] Preferably, the digital mapping table reflects the current position, movement status, idle / occupied status of all conveying units and the current operation status, queue length and availability of all working units in real time. The path planning and scheduling module is used to assign specific conveying units to each slice based on the current system status and task priority, and calculate the optimal sequence and timing of its passage through multiple working units.

[0009] Preferably, when planning a path for a delivery task, the path planning and scheduling module not only considers the shortest distance of the path, but also comprehensively evaluates the expected waiting time of the target work unit, the potential conflicts between adjacent delivery tasks, and the energy consumption index of the overall system.

[0010] Preferably, when the system control center detects a failure or performance degradation in a certain working unit, it initiates an emergency workflow to reroute the unfinished tasks of the failed working unit to other functionally equivalent normal working units, and immediately notifies the upstream conveying unit to adjust its conveying destination.

[0011] Preferably, the device has a scalable architecture that allows for the plug-and-play integration of new delivery units or work units through real-time sensing and communication networks. The system control center is able to identify newly added devices and incorporate them into the overall scheduling resource pool.

[0012] Preferably, the method of using the multi-station vertical conveyor includes the following steps: S1, the system control center is activated, and it establishes a connection with all online transmission units and working units through the real-time sensing and communication network, and constructs an initial system resource status mapping; S2, the system control center receives external production orders, decomposes them into multiple discrete conveying and processing tasks, and assigns an optimal conveying unit to each task according to the preset strategy and real-time system status. S3, for each assigned task, the path planning and scheduling module calculates one or more candidate paths based on the busy / idle status and location layout of all current work units, and selects the path with the lowest overall cost as the execution path. S4, the system control center sends a set of instructions, including path points, timing and actions, to the designated conveying unit and related working unit. The conveying unit delivers the wafers to the designated working unit in sequence according to the instructions for processing, and the units keep synchronized. S5, throughout the entire transmission and processing process, the real-time sensing and communication network continuously transmits the status data of each unit back to the system control center; S6: The system control center continuously judges whether the current plan is optimal based on the feedback real-time data. If an anomaly, conflict or higher priority task is detected, the current plan is interrupted immediately and a new scheduling instruction is generated to realize dynamic replanning.

[0013] Preferably, in step S6, when the system control center decides to replan the path, it will prioritize ensuring the continuity of high-priority tasks that are already in progress, and will compensate for the low-priority tasks affected by the adjustment by rescheduling.

[0014] Preferably, in step S6, when the real-time sensing and communication network detects or receives a fault alarm signal of a target work unit, it immediately uploads the signal to the system control center. At this time, the resource status mapping module of the system control center immediately marks the target work unit as "unavailable". Then, the path planning and scheduling module interrupts all task flows that are currently executing and whose subsequent paths contain the faulty work unit, and performs a new schedule for each affected task flow.

[0015] Preferably, the execution of the new schedule includes the following steps: querying other available work units with the same or similar processing capabilities in the system resource pool, then recalculating and selecting a new path with the best overall cost based on the query results and the current queue length and geographical location of each standby work unit, and finally issuing the new path instruction to the corresponding delivery unit and notifying the relevant work unit to update its task queue.

[0016] The beneficial effects of this invention are: 1. This invention constructs a centralized system control hub and operates it through a dynamic scheduling algorithm, enabling unified management and control of the conveying unit and various distributed work units. This significantly improves the overall operating efficiency and response speed of the multi-station vertical conveying device. The system control hub can make millisecond-level decisions and issue commands based on real-time acquired system status information, thereby ensuring seamless connection between material flow and processing, and solving the waiting and blocking problems caused by the independent operation of each unit.

[0017] 2. This invention introduces a real-time sensing and communication network, enabling the system control center to continuously acquire data fed back from the conveying unit and the working unit, including location, status, load, and even equipment health information. Based on real-time data, the device can proactively respond to various emergencies in the production process, such as equipment failure, order insertion, or fluctuations in processing time, and automatically perform task rerouting and resource reallocation, thereby significantly enhancing the reliability of the production line and reducing losses caused by unexpected downtime. Attached Figure Description

[0018] Figure 1 The diagram shown is a schematic representation of the system framework of the present invention. Figure 2 The diagram shown illustrates the workflow of this invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] Please see Figure 1 The present invention provides an embodiment of a multi-station vertical conveying device.

[0021] In this embodiment, the system control center adopts an industrial-grade server or high-performance industrial control computer hardware platform, which integrates a task management module to receive, parse, and sort task instructions from the external production management system, and assign a unique identifier and priority to each task; a resource status mapping module is responsible for maintaining a dynamically updated digital mapping table, tracking the current position, movement status, load, and health status of all conveying units in real time, as well as the operating status, task queue length, and availability of all work units; a path planning and scheduling module, relying on the A* algorithm, comprehensively considers factors such as path length, expected waiting time, risk of inter-task conflicts, and total system energy consumption when planning conveying paths, to generate the optimal task sequence and time synchronization scheme; an emergency response module can automatically activate the rerouting mechanism when a work unit failure or performance abnormality is detected to ensure task continuity; and an energy efficiency management module performs time offset scheduling for non-urgent tasks based on real-time electricity price signals and system load status to achieve economic optimization of system energy consumption.

[0022] In this embodiment, each conveying unit includes a vertical lifting mechanism (using a screw or chain lifting system) driven by a servo motor and equipped with encoder feedback, a carrier platform that can be customized according to the material type (such as pallets, clamps or shelves) and equipped with RFID or barcode identification devices for material authentication, and a local controller composed of an embedded PLC or single-chip microcomputer for receiving system instructions and feeding back status data.

[0023] In this embodiment, the working units are arranged vertically on different floors. Each unit is equipped with a local controller to receive system instructions, control equipment execution, and provide feedback on status data. It also has a material buffer area to temporarily store materials to be processed to avoid blockage. It is also equipped with status sensors (such as photoelectric sensors, pressure sensors, temperature sensors, etc.) to monitor the health status of the equipment in real time, thereby ensuring real-time communication and collaborative operation with the system control center.

[0024] In this embodiment, the real-time sensing and communication network adopts a hybrid networking method combining industrial Ethernet and Wi-Fi to construct a star or mesh topology, thereby ensuring low latency and high reliability of bidirectional data transmission throughout the system. This network is responsible for connecting the system control center, all conveying units and working units, and transmitting real-time information of the entire system, including task status, real-time location and load of conveying units, busy / idle status of working units and equipment health data. It also supports new conveying units or working units to access the network in a plug-and-play manner, which are automatically identified by the system control center and included in the unified scheduling resource pool, thereby realizing dynamic expansion and real-time coordination of system resources.

[0025] Please see Figure 2 Furthermore, the workflow of this invention will be described as follows: After the system is powered on, the system control center first performs self-check and initialization, loads modules such as task management, resource mapping and path planning, and then establishes a connection with all online devices through broadcast messages. The control center obtains the unique identifier, type, functional attributes, current location and status information (such as idle, busy, fault, power, etc.) of each device, and builds a global resource status mapping table (Digital Twin) accordingly. If a device is found to be unresponsive or in an abnormal state during the process, it is temporarily marked as "offline" or "standby" and does not participate in the initial task allocation.

[0026] The system control center receives order instructions from the upper-level production management system. Orders typically include fields such as material type, quantity, process route, priority, and delivery time. The task management module parses and decomposes the orders, generating several atomic tasks (e.g., "picking up materials → conveying to workstation A → conveying to workstation B → delivering to the storage location"). Each task is assigned a unique ID, priority label (e.g., urgent, normal, low speed), and dependencies. The resource status mapping module provides real-time information on the location, load, health status, and task queue of each conveying unit. The path planning module, based on a greedy algorithm, comprehensively considers factors such as distance, time, conflict risk, and energy consumption to allocate appropriate conveying units to each task, thereby ensuring system load balancing and efficient task execution. The greedy algorithm takes dynamic parameters such as the real-time priority of the task, the current position and status of the delivery unit, and the queue length of the target work unit as input, and uses "overall cost" (including delivery time, waiting time, conflict risk, energy consumption, etc.) as the optimization objective. In each decision step, the algorithm selects a unit from all available (idle or soon-to-be-idle) delivery units that can start the task with the minimum overall cost, without considering the potential global impact of this selection on subsequent task scheduling. The core of this strategy is to quickly generate a current optimal solution to meet the system's high requirements for real-time response. Its specific process is as follows: First, based on the task's starting location and material properties, all conveying units with the corresponding carrying capacity and in an available state are filtered out to form an initial candidate set. Then, for each conveying unit in the candidate set, the estimated comprehensive cost of performing the task is calculated, for example: The movement time is calculated based on the distance from the current position of the unit to the task starting point, the rated speed, and the acceleration. Then, the waiting time is estimated based on the current queue length and processing speed of the target working unit. Next, by querying the resource mapping table, it is determined whether the unit has spatiotemporal intersection with other task paths if it executes this task. Then, the energy consumption value is calculated based on the unit motor power, movement distance, and load. Then, the transport unit with the lowest overall cost is selected from the candidate set and directly assigned to the task. The resource status mapping table is updated immediately, and the unit is marked as "assigned". Finally, the assignment result takes effect immediately, and the task enters the execution queue without waiting for global replanning.

[0027] The path planning module first obtains real-time status data of the target work unit, including current task queue length, estimated processing time, device health, and geographical location. It then uses the A* algorithm to generate multiple feasible paths, each containing a series of waypoints and timestamps. The system calculates a comprehensive cost function for each path, with variables including path length, estimated waiting time, energy consumption, conflict probability, and priority weight. Finally, it selects the path with the lowest total cost as the final execution path, including detailed information such as action sequence, speed curve, and synchronization signal.

[0028] The control center sends task instruction packages to the local controllers of the corresponding conveying units and working units along the route through a real-time communication network. After receiving the instruction, the conveying unit starts the vertical lifting mechanism and the carrying platform, and runs according to the preset path points, speed and acceleration curves. At each node, the material identity and location are verified through RFID, QR code or visual recognition system. The working units receive task notifications in advance, prepare the corresponding tooling or programs, and immediately trigger processing, inspection or assembly operations when the material arrives. The units maintain coordinated actions to avoid collisions or waiting, thereby ensuring continuous and uninterrupted material flow.

[0029] The sensors built into the conveying unit and the working unit (such as encoders, photoelectric switches, force sensors, and temperature sensors) collect real-time operating data, including position, speed, load, energy consumption, equipment temperature, and vibration amplitude. The local controller preprocesses and packages this data, and sends it to the system control center via wired or wireless network through Pub / Sub mode or periodic reporting mechanism. After receiving the data, the resource status mapping module dynamically updates the global mapping table, recording the execution progress of each task, equipment utilization, abnormal events (such as timeouts, deviations, and fault codes), etc. At the same time, the data storage module writes historical data into the time-series database for subsequent analysis and optimization.

[0030] During operation, the control center continuously compares the planned and actual execution status. If path deviation, equipment failure, task timeout, insertion of emergency orders, or sudden changes in system load are detected, a replanning process is triggered: the task management module suspends the affected tasks, the path planning module recalculates feasible paths, and the resource mapping module updates the equipment status. During replanning, the continuity of high-priority tasks is prioritized, and low-priority tasks that are interrupted are compensated and scheduled (such as early execution, allocation of spare resources, or path adjustment). After the new instructions are generated, they are immediately sent to the relevant units. The unit's local controller supports hot switching of instructions, and can switch to the new task flow without a complete shutdown, thereby achieving dynamic response and system recovery with minimal interruption time.

[0031] Furthermore, in this embodiment, the fault handling process is described as follows: When the system is running, if the real-time sensing and communication network detects or receives a fault alarm signal (such as overload, over-temperature, communication interruption, mechanical jamming, etc.) actively reported by a target work unit, the signal will be immediately uploaded to the system control center. The resource status mapping module will update the status of the target work unit from "available" to "unavailable" or "faulty" within a millisecond response time, and temporarily remove it from the current schedulable resource pool. At the same time, it will record the fault type, occurrence time and estimated recovery time (if it can be estimated). Then, the path planning and scheduling module will automatically start the emergency workflow, interrupt all task flows that are currently executing and whose subsequent paths contain the faulty work unit, and perform new scheduling for each affected task flow.

[0032] The process first quickly queries other available work units with the same or similar processing capabilities in the global resource pool. Then, based on the real-time acquisition of the current task queue length, geographical location, equipment health status, and energy consumption status of each standby work unit, a multi-objective optimization algorithm is used to recalculate and select a new path with the best overall cost. An instruction containing the new path point, timing synchronization information, and action instructions is generated. Finally, the system sends the new path instruction to the corresponding conveying unit and notifies the relevant work unit to update its local task queue and production plan.

[0033] Furthermore, the present invention provides embodiments: This embodiment uses a multi-story factory of a large personal hygiene products manufacturer as an application scenario. This company produces various disposable absorbent pad products, including sanitary napkins, diapers, nursing pads, and breast pads. The product structure typically includes multiple layers of materials such as an absorbent core, a non-woven fabric surface, and a leak-proof bottom film, and production is completed through processes such as lamination, molding, cutting, and packaging. The factory has four floors, each with workstations for different functions. Traditional conveying methods are inefficient and lack flexibility, unable to meet the agile production needs of multiple varieties and small batches.

[0034] This embodiment integrates a task management module, a resource status mapping module, a path planning and scheduling module, and an emergency response module. The task management module receives production orders from the upper-level MES. The orders include information such as product model, quantity, process route (e.g., "composite → molding → cutting → packaging"), and priority. The resource status mapping module dynamically maintains a digital mapping table and tracks the status of all conveying units and working units in real time. The path planning and scheduling module uses an improved A* algorithm to comprehensively consider path distance, target workstation queue length, task priority, and total system energy consumption when planning paths, allocating appropriate conveying paths for each batch of absorbent sheet products.

[0035] This embodiment has a total of 4 conveying units. Each unit includes a precision vertical lifting mechanism driven by a servo motor and a carrying platform. The platform is equipped with a vacuum suction cup to adapt to the soft and easily deformable characteristics of products such as sanitary napkins and diapers, and to avoid wrinkles or damage during the conveying process. At the same time, the carrying platform integrates visual recognition to identify product batch information. Each conveying unit is equipped with an embedded PLC as a local controller, which is responsible for receiving central instructions and feeding back real-time position, load and operating status.

[0036] The work units are distributed vertically across four floors: Floor 1 is the raw material compounding and initial processing station, equipped with a multi-layer material compounding machine; Floor 2 is the precision forming and cutting station, equipped with a CNC die-cutting machine and an ultrasonic welding machine; Floor 3 is the quality inspection station, equipped with a high-speed vision inspection system for detecting product integrity, surface defects, and dimensions; and Floor 4 is the packaging and palletizing station, equipped with an automatic packaging machine and a robotic arm. Each work unit has a local controller and a material buffer area, and is equipped with photoelectric sensors, temperature sensors, and vibration sensors for real-time monitoring of equipment status and communication with the system control center.

[0037] After the system starts up, it scans and registers all devices, builds a global resource status mapping table, and then receives a production order for "5000 ultra-thin sanitary napkins" from the MES. The task management module decomposes this into a series of atomic tasks (such as "material picking → conveying to the composite station → conveying to the molding station → conveying to the testing station → conveying to the packaging station") and assigns a priority to each task.

[0038] The path planning module dynamically calculates the optimal path for each batch of products based on the real-time system status (such as a long queue at the 2nd-floor forming station and idle workstations at the 3rd-floor testing station). It may decide to prioritize some batches to the 3rd-floor testing station to avoid congestion. The conveying unit picks up the products with vacuum suction cups according to instructions and smoothly conveys them to each workstation according to the planned path. At each workstation, the batch is verified by the vision system.

[0039] During execution, if a visual inspection device on the 3rd floor triggers an alarm due to lens contamination, the control center immediately marks it as "unavailable" and activates the emergency response module. All affected tasks are dynamically rerouted to another normally functioning inspection device on the 3rd floor, and the relevant transport units are notified to update the destination.

[0040] The implementation of this device enables efficient, flexible, and damage-free vertical transport of various disposable absorbent sheet products in multi-layer production lines, improving overall production efficiency and system reliability.

[0041] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A multi-station vertical conveying device, characterized in that, Including: The system control center is used to process global task instructions, allocate system resources, calculate the optimal transmission path, and generate a set of synchronization control instructions. Multiple conveying units are controlled by the system control center. Each conveying unit includes an independently controllable vertical lifting mechanism and its supporting platform for conveying the wafers in the vertical direction. Multiple work units, each equipped with a local controller and communicating in real time with the system control center, are used to perform predetermined processing, inspection, assembly or storage operations on the wafers delivered by the conveyor unit; A real-time sensing and communication network connects the system control center, all conveying units, and all working units, enabling bidirectional transmission of system-wide information, including task status, real-time location and load of conveying units, busy / idle status of working units, and equipment health data.

2. The multi-station vertical conveying device according to claim 1, characterized in that: The system control center includes a task management module, a resource status mapping module, and a path planning and scheduling module. The task management module is used to receive, parse, and queue task instructions issued by the external production management system, and assign a unique identifier and priority to each task. The resource status mapping module is used to maintain a dynamically updated digital mapping table.

3. The multi-station vertical conveying device according to claim 2, characterized in that: The digital mapping table reflects in real time the current position, movement status, idle / occupied status of all conveying units and the current operation status, queue length and availability of all working units. The path planning and scheduling module is used to assign specific conveying units to each slice based on the current system status and task priority, and to calculate the optimal sequence and timing of its passage through multiple working units.

4. The multi-station vertical conveying device according to claim 1, characterized in that: When planning a path for a delivery task, the path planning and scheduling module not only considers the shortest distance of the path, but also comprehensively evaluates the expected waiting time of the target work unit, the potential conflicts between adjacent delivery tasks, and the energy consumption index of the overall system.

5. The multi-station vertical conveying device according to claim 1, characterized in that: When the system control center detects a failure or performance degradation in a certain working unit, it initiates an emergency workflow to reroute the unfinished tasks of the failed working unit to other normally functioning units with equivalent capabilities, and immediately notifies the upstream transport unit to adjust its transport destination.

6. The multi-station vertical conveying device according to claim 1, characterized in that: The device has a scalable architecture, which allows for the plug-and-play integration of new delivery units or work units through real-time sensing and communication networks. The system control center can identify newly added devices and incorporate them into the overall scheduling resource pool.

7. A method of using a multi-station vertical conveying device, employing the multi-station vertical conveying device according to any one of claims 1-6, characterized in that, It includes the following steps: S1, the system control center is activated, and it establishes a connection with all online transmission units and working units through the real-time sensing and communication network, and constructs an initial system resource status mapping; S2, the system control center receives external production orders, decomposes them into multiple discrete conveying and processing tasks, and assigns an optimal conveying unit to each task according to the preset strategy and real-time system status. S3, for each assigned task, the path planning and scheduling module calculates one or more candidate paths based on the busy / idle status and location layout of all current work units, and selects the path with the lowest overall cost as the execution path. S4, the system control center sends a set of instructions, including path points, timing and actions, to the designated conveying unit and related working unit. The conveying unit delivers the wafers to the designated working unit in sequence according to the instructions for processing, and the units keep synchronized. S5, throughout the entire transmission and processing process, the real-time sensing and communication network continuously transmits the status data of each unit back to the system control center; S6: The system control center continuously judges whether the current plan is optimal based on the feedback real-time data. If an anomaly, conflict or higher priority task is detected, the current plan is interrupted immediately and a new scheduling instruction is generated to realize dynamic replanning.

8. The method of using the multi-station vertical conveying device according to claim 7, characterized in that: In step S6, when the system control center decides to replan the path, it will prioritize ensuring the continuity of high-priority tasks that are already in progress, and will compensate for the low-priority tasks affected by the adjustment by rescheduling them.

9. The method of using the multi-station vertical conveying device according to claim 7, characterized in that: In step S6, when the real-time sensing and communication network detects or receives a fault alarm signal of a target work unit, it immediately uploads the signal to the system control center. At this time, the resource status mapping module of the system control center immediately marks the target work unit as "unavailable". Then, the path planning and scheduling module interrupts all task flows that are currently executing and whose subsequent paths contain the faulty work unit, and performs a new schedule for each affected task flow.

10. The method of using the multi-station vertical conveying device according to claim 9, characterized in that: The execution of the new schedule includes the following steps: The system queries other available work units with the same or similar processing capabilities in the system resource pool. Based on the query results and the current queue length and geographical location of each standby work unit, it recalculates and selects a new path with the best overall cost. Finally, it issues the new path instruction to the corresponding transport unit and notifies the relevant work unit to update its task queue.

Images