Artificial Intelligence Monitoring System Using ZigBee Wireless Network Technology in Warehousing and Logistics Innovation and Economic Cost Management
The study expects to reduce the development cost and energy consumption of the storage environment monitoring and improve the efficiency of the storage monitoring environment. Firstly, the storage environment monitoring system is designed in detail using ZigBee. It includes the design of terminal nodes, router nodes, coordinator nodes, and upper computer software. Here is a detailed description of the process of establishing a communication network for three different types of devices in the ZigBee network. The address allocation mechanism of the ZigBee network is introduced. Then, through the application of activity-based costing, the cost reduction is achieved as expected. Finally, the evaluation model of the storage logistics cost level is established, and the various reasons that affect the storage logistics cost are analyzed. The completed warehouse environment monitoring system is a tree topology network composed of a ZigBee coordinator, several ZigBee routing nodes, and terminal devices through the above design. The coordinator in the network is responsible for establishing the network, and it can communicate with the host computer through the serial port. The terminal node collects environmental parameter information and feeds the data back to the ZigBee coordinator through wireless multihop. The ZigBee routing node is mainly used to expand the range of the network to achieve the purpose of wireless monitoring of the entire storage environment. This provides a reference for the combination of wireless communication technology and sensor technology.
1. Introduction
Nowadays, the rapid development of China’s productivity has driven the development of the logistics industry. Logistics has become an important cornerstone and guarantee of economic growth. It is known as the “third source of profit” since it reduces material consumption and increases labor-driven output [1]. The logistics industry is an industry that integrates warehouse storage, logistics transportation, and information and data integration. It is one of the measurement standards for weighing the strength of an enterprise or even a country. Among them, the construction of warehouse storage is a crucial part of contemporary logistics, and its importance is increasing day by day. Warehouse storage is now a key content and an indispensable part of logistics management [2].
In the technical field of warehouse storage, intelligent warehouse storage is the inevitable result of development [3]. Integrating smart warehouse storage and wireless communication also moves warehouse storage toward a more philosophical direction [4]. A vital part of the unmanned warehouse storage is the internal environment control system of the warehouse storage. The internal environment of warehouse storage can be changed through monitoring to ensure the safety and quality of the items stored in the warehouse [5].
The logistics level of all walks of life is improving. The management and control of the internal situation of warehouse storage also need to keep up with the development of the legal person of the times. In this process, the most important thing is the critical data stored in the warehouse, such as humidity and temperature. They greatly influence the quality and consumption of the items placed in the warehouse [6]. At present, the monitoring and control of the internal conditions of warehouses in China are still in manual inspection and wired inspection. These two storage management methods have shortcomings: (1) staff inspections and workers’ labor costs are high. It is time-consuming and laborious, and the efficiency is not very high. The accuracy of the data is not high enough [7]. (2) Wired monitoring accurately reflects the warehouse storage dynamic data on internal conditions. The intricacies of wired detection lines restrict the flexibility of wired monitoring. The maintenance and management costs are also high [8, 9]. Here, an artificial intelligence (AI) monitoring warehouse storage system is built using ZigBee wireless network technology. ZigBee uses energy to monitor various metrics and data stored in warehouses dynamically. If an abnormality is detected in the internal environmental indicators, the sensor will alert. The alarm information can notify the upper computer operator and deal with it in time. ZigBee is low-cost, has low energy consumption, and is easy to install. ZigBee is used to build an AI monitoring system for the internal environment of the warehouse, which can skillfully deal with a series of problems such as complex wires, insufficient flexibility, and relatively high cost in the previous warehouses. It has enormous advantages that warehouse storage management did not have before. This provides more advanced management technology for the logistics management system and contributes to the improvement of the logistics management model.
2. Materials and Methods
2.1. ZigBee Technology Introduction
It was discovered that bees communicate through Z-flying, so ZigBee is named after the communication method of bees [10]. In short, ZigBee is a highly reliable wireless data transmission network, similar to Code Division Multiple Access (CDMA) and Global System for Mobile Communications (GSM) networks. The ZigBee data transmission module is similar to the mobile network base station. The communication distance is from the standard 75 m to several hundred meters and several kilometers, supporting unlimited expansion. ZigBee is a wireless data transmission network platform composed of up to 65535 wireless data transmission modules. Each ZigBee network data transmission module can communicate with each other in the whole network range. The distance between each network node can be extended infinitely from the standard 75 m. Unlike CDMA or GSM networks, the ZigBee network is mainly established for industrial data transmission for field automation control. Therefore, it must have the characteristics of simplicity, ease of use, reliable operation, and low price. The mobile communication network is mainly established for voice communication, and the value of each base station is generally more than one million yuan. Each ZigBee “base station” costs less than 1000 yuan. Each ZigBee network node can be used as a monitoring object itself and automatically transfer data from other network nodes. For example, the connected sensors directly collect and monitor data. In addition, each ZigBee network node can wirelessly connect with multiple isolated subnodes that do not undertake the task of network information transfer within the range covered by its signal [11, 12]. With the rapid development of ZigBee technology, it has been widely used in medicine, plant protection, and various intelligent equipment. With the advancement of the Internet of Things (IoT), ZigBee technology is used in all walks of life [13, 14].
There are three types of network topology in ZigBee technology: star topology [15], tree topology, and mesh topology [16]. Figure 1 shows the type topology.
In Figure 1, the star topology is radial as a whole. The only node core of the structure is the coordinator. This type of network-related configuration is constructed through the coordinator. Many terminal systems on the periphery communicate directly with the coordinator. The characteristic of this model is that there is no complicated system, not much information, and the price and cost are not high. Figure 2 shows the tree topology.
In Figure 2, it is a tree topology. It is tree-shaped, and the only node core is the coordinator. Figure 3 shows the mesh topology.
Figure 3 shows the mesh topology. It is composed of a coordinator and multiple terminals or routers. The problem in the structural path is automatically identified by the wrong direction to communicate. It is more sensitive and can be better presented in performance and safety. However, because of these advantages, it has many channels and is complicated, and the amount of data saved is relatively large. Compared with the two structures shown in Figures 1 and 2, the structure shown in Figure 3 consumes more energy, and the construction and daily maintenance costs are higher. Therefore, the agency is used in situations where the environment is more complex, and the degree of spread is relatively large [17]. Table 1 shows the comparison between ZigBee wireless network technology and other network technologies.
|
|
||||
| Name | Bluetooth | Wi-Fi | UWB | ZigBee |
|
|
||||
| Transfer speed | 1 Mbps | 11-54 Mbps | 53-480 Mbps | 100 Kbps |
| Communication distance | 20-200 m | 20-200 m | 0.2-40 m | 2-20 m |
| Frequency band | 2.4 GHz | 2.4 GHz | 3.1 GHz 10.6 GHz |
2.4 GHz |
| Safety | High | Low | High | High |
| Power consumption | 20 mA | 10-50 mA | 10-50 mA | 5 mA |
| Cost | 2-5$ | 25$ | 20$ | 5$ |
| Main application | Communication, automotive, IT, multimedia, industrial, medical, education | Wireless Internet, PC, PDA | High-fidelity video, wireless hard drive, etc. | Wireless sensor, medical |
|
|
||||
In Table 1, compared with other network technologies, ZigBee wireless network technology, although the transmission rate is slower, has more advantages in power consumption and cost. Compared with Wi-Fi and UWB (ultra-wideband) network technology, ZigBee wireless network technology is more compact and flexible in use and has high insecurity. Meanwhile, ZigBee wireless network technology can be combined with wireless sensors. Therefore, its use effect in the process of logistics management is better.
2.2. Overview of Activity-Based Costing and Comparison with Traditional Costing
Activity-based costing is described by “Activity consumes resources, products consume Activity”. Activity-based costing includes attributing resources and the motivation of movement. Establishing the core of activity is the main content of this principle. Once each type of motivation is selected, the resource cost can be accurately and reasonably used in the final product. Various types of motivation can be understood according to different activities. The collection is called the cost library. This is the activity center [18].
2.2.1. Different Cost Allocation Methods
If the product cost is divided into direct or indirect costs, it includes the direct or indirect costs that can directly have the finished product. The traditional cost method generally controls indirect costs using direct costs and uses an even distribution. However, activity-based costing separates each link from time to time, establishes an activity center for each link, and controls indirect costs using the amount of activity required for each type of product. Activity-based costing is controlled by disassembling cost drivers. It can also analyze value-added and non-value-added activities and understand the impact of each activity link on product costs. These are superior to the traditional cost method [19].
2.2.2. Different Cost Accounting Objects
In the past, the cost calculation method was used to account for resources, and activity-based costing was used to account for activity. And the cost method was calculated through the depletion and loss of materials. What is estimated is the target salary, data cost, etc., which can be recorded in the ledger. However, activity-based costing uses the number of activities used by the product, and movement controls the cost through the composition of resources. It can decompose resources into detailed application steps or active management. It is like using orders to process activity. The number of applications of orders is carried out by order managers, who are related to the company’s salary and other information. Therefore, activity-based costing dictates wages to product costs and indirect costs, that is, wages, using the frequency of order usage. Thus, the goals between activity-based costing and previous costing methods are different [20].
2.2.3. Different Cost Management Purposes
All costs consumed by the company are classified and summarized. This is the traditional cost method to calculate the total cost. In this way, the consumption expenses of the company and products can be translated, reflecting the total cost and final cost of individual products. Activity-based costing is different. It can feedback the company’s price or product and reflect the various resources consumed in other activities. It feeds back the result and provides around the consumption of its process. Activity-based costing can better reflect the causal relationship between process volume and cost and various types of consumption [21].
2.3. Applicability of Activity-Based Costing
In an enterprise, each department performs its duties. The direct cost of materials and labor accounts for a small proportion of the total cost of the enterprise. From the cost analysis of a particular month, the storage cost of a company’s warehouse includes the entrance department, packaging department, exit department, and warehouse control department. Among them, if the management fee only comes from the administrative department, warehouse storage costs account for 76.9% of the total cost. However, when products are put into storage, they are usually many products of different types, which need to be inspected by the admission department. The task of the packaging department is to sort and pack. The library control department is deployed in the entire system, so it is not easy for each department to allocate costs directly. Only the cost of the field department is easy to grasp. One appearance is the labor cost of appearance. Then, according to this, the labor cost is the material and labor cost of the field department and the packaging department, and the indirect cost includes the remaining departments of the enterprise. According to the above accounting, direct costs are only about 30%. When the indirect cost is relatively high, activity-based costing is to allocate resources according to the activity motivation. Compared with the traditional cost method, it is more scientific and reasonable to distribute only the working hours of people and machines. Meanwhile, the company has also reduced the overall accounting system, and the cost calculation method remains unchanged.
2.4. Construction of the Evaluation Model of Warehousing Logistics Cost Level
Combined with the proportion of the warehousing logistics cost factor in the total logistics cost, the weighting standard of the logistics cost factor in this industry has been established, which is and .
where represents the weight matrix. The Likert scale is divided into three steps. First select the impact factors in the system for analysis. Through the study of each influencing factor, the results are as follows: (1) higher cost, (2) large cost, (3) medium cost, (4) small cost, and (5) less cost.
Then, give a score to the impact factor under the test; for example, the higher the cost, the smaller the corresponding price of 5, 4, 3, 2, and 1. The company’s warehousing logistics cost level is divided into five groups. The related scores are 5 points, 4 points, 3 points, 2 points, and 1 point. Use the letter to represent the score and the letter to represent a vector.
Finally, the measured object’s evaluation scores for the influencing factors in warehousing logistics are calculated, and the average value of the corresponding project scores is calculated. Make it the score of the logistics cost factor, and get the total score of the individual attitude of each logistics cost factor. Meanwhile, record the number of people corresponding to each score and then calculate the probability of each logistics cost factor to obtain the score. ; that is, the probability of each logistics cost factor in the test item is , and =1. The logistics cost factor Likert scale has been constructed.
The probability value of the company’s total cost in five different grades can be obtained from the logistics cost control level, which is represented by the letter .
3. Design of a Warehouse Environment Monitoring System Using ZigBee
3.1. Terminal Node Design
The terminal nodes are placed in the warehouse storage [22], and the specific locations are strictly following the standard requirements. This part is the most basic construction of the entire structure. Firstly, use the temperature and humidity sensor (DHT11) and the gas sensor (MQ-2) in the warehouse to collect various indicators in the storage. Secondly, the data is managed by the processor and processed accordingly. Again, the wireless signal receiver and receiver are handed over to the router node. Finally, the information can be guaranteed to be transmitted to the coordinator node. After the above experience, the coordinator built access to the ZigBee wireless network [23, 24]. Figure 4 is a block diagram of the hardware structure of the terminal node.
In Figure 4, the column parameters of the terminal node are designed. It can be used according to the required functions and performance. The power supply uses batteries to provide power.
3.2. Design of the Router Node
In constructing the entire structure, the router node is responsible for communicating and relaying. The coordinator communicates with the monitoring center and the terminal serial port, and the terminal node is placed in the warehouse storage. Therefore, when the distance between the terminal node and the coordinator node exceeds the maximum threshold of the communication distance, the router node will continue the communication relay [25]. The router node does not process arbitrary information. It is only for transmitting the information. Like the terminal node, the router node must ensure that it builds the ZigBee system within the feasible range of communication before the task is carried out. Of course, in an unconventional scenario, both nodes are in the communication range of the coordinator. The router node will fail, which means that the terminal node will communicate with the coordinator node instead of interacting with the router node. Usually, the router node is built outside the warehouse storage, and it is ensured that each warehouse environment has at least one router node. Figure 5 is a block diagram of the hardware structure of the router node.
In Figure 5, compared with the terminal node, the router node usually processes the RF information and the power supply part and does not collect the data information. Because the router node ensures the radius of communication, it is in the information system of the terminal node’s signal reception and release. The power source of the router node is direct current instead of guaranteeing power through dry batteries.
3.3. The Design of the Monitoring Software of the Host Computer
The monitoring and control system of the host computer can provide dynamic information and data to assist the staff in understanding the interior of the storage environment. It can also deal with accidents in the internal situation of the warehouse in time [26]. The monitoring carried on the monitoring control terminal has the following functions: (1) it can receive the information sent by the coordinator through the serial port and combine the information to process and analyze the information, (2) it can provide the processed data to work through specific digital and graphical information personnel, (3) it can distinguish according to the concentration of the gas and send out an alarm if there is an abnormality, and (4) it can save and process data information according to the database. Figure 6 is a block diagram of the functional design of the software system.
Figure 6 shows the design of the host computer monitoring and control system, which includes the collected information display part, the temperature and humidity trend part, the past information part, and the corresponding configuration part. The assembled information display part is represented by four types of information: power, humidity, temperature, and gas concentration. The temperature and humidity trend part is the corresponding change curve, which allows the staff to make corresponding actions in combination with the related data [27, 28]. The past information part stores the previous information, including the changing curve. The corresponding configuration part has exposure adjustment and information integration methods, including manual and automatic integration.
3.4. The Design of the Upper Computer Database
The collected information and environment-related parameters are stored and processed in the database, 5ensuring the continuity of the information and protecting the data from risk [29]. Process the data that needs to be preprocessed, including analyzing abstractions and building models. There will be a lot of data in the system being built, including the bottom-level data and environmental information.
In Figure 7, the specific information involved in the system is analyzed and abstracted, and a model is constructed. The information involved in this system generally includes two categories: basic node information and environmental data information. According to the E-R diagram, the available relationship modes can be divided into basic node information (node ID, node power) and environmental data information (temperature data, humidity data, and gas data). In summary, this study achieves high precision, miniaturization, low power consumption, intelligent and other specific requirements through the design of terminal nodes, terminal node sleep mechanism and power query router nodes, host computer monitoring software, host computer control software power early warning mechanism, and host computer database. This provides an essential reference for the combination of wireless communication technology and sensor technology. The combination of ZigBee wireless network technology and wireless sensor technology can improve the functional characteristics of ZigBee and expand the main functions of ZigBee technology. The main features are control, energy savings, convenience, and safety. Firstly, the control function is that the combination of ZigBee wireless network technology and wireless sensor can increase the process of both. This joint control function can be used in a variety of situations. Logistics management is one of the occasions where it can be applied. The energy-saving feature means that the combination of ZigBee technology and wireless sensing technology can perform cost detection and management of the factors it controls to provide the best energy-saving method. The convenient feature means that the combination of ZigBee wireless network technology and sensors can be installed without any wires and operated and managed by a remote control system. The security feature means that after ZigBee is combined with wireless sensors, it can detect the status of its monitoring factors and monitor and alarm its abnormality, thereby improving its security. Therefore, the proposed method has significant advantages.
4. Result
4.1. Implementation of the Coordinator Node
The function of the coordinator node design is to form a ZigBee network, allow other nodes to join the network, manage network nodes and assign addresses, wirelessly receive data sent by other nodes, and upload the serial port to the host computer for display. Figure 8 shows the workflow of the coordinator node.
In Figure 8, (1) after the coordinator node is powered on, a series of initialization tasks can be completed. (2) The coordinator node performs channel energy detection, selects a proper channel to form a ZigBee network, and sets the unique PANID identifier and 16-bit network short address. Generally, 0x0000 is chosen as its fixed address by default, and the network creation is now complete. (3) When other nodes within the communication range of the coordinator node search for the ZigBee network, they will initiate a request to join the network to the coordinator node. If the node information meets the requirements, the coordinator node will allow the node to join its network. Then, it can assign a 16-bit network short address to the node and configure the relevant network parameters, and the coordinator node records the joining of the network node in the neighbor node list. (4) In the application scenario of this system, the data finally converges to the coordinator node through the router node. The coordinator node transmits it to the monitoring center computer. Through the host computer monitoring software, the data is displayed.
4.2. Implementation of Router Nodes
Figure 9 presents the workflow of the router node. Before powering the router, adjust the configuration parameters (coordinator, router, end device, etc.) in the system. In other words, the logical type of the system is output from the storage device, and the logical type of the system is adjusted to ZG_DEVICETYPE_Router. The kind of system startup will be changed to ZG_DEVICETYPE_Router after the ZigBee router is powered. It resets every part of its hardware system and software system. Then, the scan DEFAULT_CHANLIST is set to the specified position. Combine the function ZStatus_t NLME_NetworkDiscoveryRequest() to turn on the search helper. If it is integrated into the coordinator, the router will start the operation ZStatus_t NLME_OrphanJoinRequest() to apply for joining the network. Meanwhile, there will be some feedback from the network assistant. After the network assistant receives the router’s application, if it agrees to access the network, the router will get the address assigned by the coordinator. And then, the router will wait to accept the instruction of the terminal device to join the network. At this time, there will also be data packets from the ZigBee coordinator and ZigBee terminal equipment at the same time. The corresponding code can be added according to actual needs so that both the router and the terminal node can collect the internal resources of the warehouse storage. The SennorApp_ProcessEvent processing function can be used in the application layer.
4.3. Implementation of Terminal Nodes
The SennorApp_ProcessEvent processing function is added to the application layer in the terminal stage. The node has information reception and the ability to transmit wireless information. Figure 10 shows the workflow of the sensor node.
In Figure 10, the terminal node will find and apply to join the network immediately after it is powered. Execute the function zb_StartRequest() in the ZB_ENTRY_EVENT event. The newly entered node must combine with the router and coordinator in the existing network before successfully joining the network. The routers and coordinators in the communication range will respond to the data packets applied for access to the network and give the node a dedicated network address. Enter the low power consumption mode of “standby-working state (collecting data) standby”. Collecting and transmitting data are achieved through the void myApp_Measure() function.
4.4. Network Address Allocation Plan
According to the actual needs of the system establishment, to save costs, a tree topology network is selected. Each node can only communicate with its previous or next node in the tree-shaped ZigBee network. In this way, it is convenient for information to be transmitted through tree-shaped lines, and network addresses are allocated in the form of preallocation. Each system that has just joined has its network address. Figure 11 shows the network address allocation.
The ZigBee coordinator or router assigns the network address to the next node according to the initially set data and the established algorithm. The related configurations are as follows: the maximum depth of the network depends on Lm, which limits the length of the network from a physical point of view. After the coordinator builds the network again, the network address is set to 0, and the depth is recorded as . Its next node depth is recorded as 1. Analyzing the maximum capacity of the next router depends on Rm. The maximum ability of the communication next node depends on Cm.
4.5. Application Analysis of the Evaluation Model of Warehousing Logistics Cost Level
The weight of storage cost and transportation cost is 0.3 through the loading of t5he model. The weight of handling costs and distribution costs is 0.15. The packaging cost weight is 0.1. Among them, all the values represent weights and have no meaning. In the probability map of the impact factor, 3 points are the most, indicating that the logistics management cost is moderate, but there is no advantage. The results are shown in Figures 12 and 13.
In Figures 12 and 13, the logistics process is evaluated by warehousing costs, handling costs, packaging costs, transportation costs, and distribution costs. Firstly, warehousing and transportation costs are the two most expensive aspects of logistics management. A packaging cost is one of the least costly aspects of the entire logistics management process. The highest score of logistics management is 3 points through the evaluation of the score method. In logistics management, logistics management still has room for improvement, and more optimization is needed. The application of the evaluation model of the warehousing logistics cost level is analyzed. The results show that the control of logistics costs is not unilateral. Enterprises should pay attention to the relationship between warehousing, logistics, and cost according to the “antinomy” (the antinomy refers to two theories or doctrines on the same object or problem, although valid but contradictory). It would help to focus on the overall impact, not on a single factor. Therefore, by establishing the evaluation model of warehousing logistics cost level, it can better reflect the general situation of enterprises in the control of warehousing logistics cost and provide a good reference for enterprises to formulate logistics cost control countermeasures.
5. Conclusions
ZigBee technology and its characteristics are introduced. Its architecture is described, and the system is designed in detail, including detailed design of terminal nodes, router nodes, coordinator nodes, and upper computer software. Establishing a communication network by the three devices in the ZigBee network is described in detail. The address allocation mechanism of the ZigBee network is introduced, and activity-based costing is applied to achieve the expected cost reduction. The evaluation model of the storage logistics cost level is established, and the various reasons that affect the storage logistics cost are analyzed. The system is designed in detail, but it needs further exploration. These include (1) realizing mesh topology. The mesh topology can accommodate more nodes than the cluster tree topology. The mesh topology has self-organization and self-healing functions and high reliability. Using a mesh network can effectively expand the network range. (2) The knowledge of hardware needs to be further studied. This helps to solve problems in time when the equipment is debugged. In the future, after learning, the system will be improved so that the storage environment can be accurately monitored.
Data Availability
The data used to support the findings of this study are included within the article.
Conflicts of Interest
The authors declare no conflicts of interest.