**3. Results**

In this section, a brief explanation of the results is provided. Section 3.1 presents the process of data collection and dataset information. Section 3.2 presents the performance evaluation of the proposed system. Section 3.3 presents smart manufacturing challenges and opportunities.

#### *3.1. Implementation Environment*

The implementation of the proposed system structure and environment is presented in this section. Table 3 summarizes the purposed system experimental setup. All the experiments were done on an Intel(R) Core(TM) i7-8700 @3.20 GHz processor with 32 GB memory. Moreover, the docker environment was processed in the 18.06.1-ce version, and the container configuration in the virtual machine was processed based on the docker composer 1.13.0 version. The Hyperledger Fabric framework project is from the Linux Foundation.

**Table 3.** Development environment of the proposed system.


Figure 5 shows the operation of the transaction process function. For improving the assets and participants, create, delete, update, and other functions were defined in the blockchain network. The functions of the transaction processor were implemented in JavaScript and defined as a smart contract. The specified ShareRecord function is used to update the manufacturing records based on the events and registry.

To control the domain model elements, the access control language (ACL) is needed. ACL provides the ability to define rules to specify the roles and users, which are authorized to make changes in the business network domain. Figure 6 shows the ACL rules defined in this network that give participants access to make changes in the network.

**Figure 6.** Access control definition in the proposed manufacturing system.

#### *3.2. Dataset Management*

The smart manufacturing system's data increase in volume based on the traditional algorithms' ability, mostly when the user wants to extract useful information from the collected dataset. High sample volume in a large dataset, when the records are not similar, needs the consolidation and isolation algorithms for implementation and knowledge utilization. In this research, the data were collected from various sources related to IoT; the production equipment was collected from various sensors to monitor the product in real-time—e.g., the built-in sensors measured, monitored, and reported the status of manufacturing equipment and product based on the temperature, humidity, pressure, etc. Figure 7 shows the data-driven process in smart manufacturing.

**Figure 7.** Data-driven process.

Table 4 presents the configuration of IoT devices and sensors for real-time data collection. During the smart manufacturing (SM) cycle, the IoT devices are located in the main areas of manufacturing resources at various levels, e.g., machines, factories, etc. The radio-frequency identification (RFID) tags are mainly configured enough in practical documents to report important machines' quality, design, and production procedures in the manufacturing process.

**Table 4.** IoT device configuration information for data usage in smart manufacturing.

