A small and medium enterprises (SMEs) manufacturing platform aims to perform as a significant revenue to SMEs and vendors by providing scheduling and monitoring capabilities. The optimal job shop scheduling is generated by utilizing the scheduling system of the platform, and a minimum production time, i.e., makespan decides whether the scheduling is optimal or not. This scheduling result allows manufacturers to achieve high productivity, energy savings, and customer satisfaction. Manufacturing in Industry 4.0 requires dynamic, uncertain, complex production environments, and customer-centered services. This paper proposes a novel method for solving the difficulties of the SMEs manufacturing by applying and implementing the job shop scheduling system on a SMEs manufacturing platform. The primary purpose of the SMEs manufacturing platform is to improve the B2B relationship between manufacturing companies and vendors. The platform also serves qualified and satisfactory production opportunities for buyers and producers by meeting two key factors: early delivery date and fulfillment of processing as many orders as possible. The genetic algorithm (GA)-based scheduling method results indicated that the proposed platform enables SME manufacturers to obtain optimized schedules by solving the job shop scheduling problem (JSSP) by comparing with the real-world data from a textile weaving factory in South Korea. The proposed platform will provide producers with an optimal production schedule, introduce new producers to buyers, and eventually foster relationships and mutual economic interests.
Production scheduling can maximize customer satisfaction by increasing the efficiency of the shop floor and minimizing production time in Industry 4.0. The increased efficiency brings the cost reduction effect due to minimizing labor cost and delivery time [
To be specific, in the case of textile manufacturing, this industrial transformation hugely impacts the industry. The textile industry in South Korea heavily depends on SMEs. According to the National Statistical Office in Korea, 99.97% of the textile garment industry was owned by SMEs in 2018. The large companies occupy the remaining percentage. At the same time, this trend has continued until recently. Without the manufacturing software, the scheduling tasks are entirely up to workers, that their scheduling is not as efficient as the software can schedule. In this sense, SMEs are suffering from severe economic difficulties. A solution should be proposed to overcome the limitations of the economic difficulties of SMEs mentioned above. Therefore, this paper proposes a method to solve SMEs’ limitations in economic difficulties and overcome the disadvantages of digital transformation. This will bridge the gap between large companies and SMEs by minimizing the flow of orders in textile manufacturing.
In this sense, in the study, we used textile manufacturing SMEs data in Korea to test the genetic algorithm (GA) model to solve the job shop scheduling problem (JSSP) and obtain the optimal scheduling result. GA is a heuristic searching algorithm inspired by natural selection and genetic nature [
Although GA is optimal for solving the JSSP in the platform, it takes time to schedule with massive JSSP instances. Various methods have been studied recently to overcome the shortcomings of GA. A representative method is to apply a machine learning-based method [
In this study, a scheduling system for the SMEs manufacturing platform is proposed and is expected to provide users with fast and efficient scheduling ability and reflect the real-world practice of the existing field. The job shop scheduling engine has an algorithm for solving the JSSP to get the optimized scheduling result. The rest of the paper is presented as follows. Section 2 introduces knowledge of the manufacturing platform and JSSP. Section 3 briefly introduces the JSSP model of the platform system. Section 4 simulates the JSSP model of the platform with the data of textile manufacturing owned by SMEs in South Korea and describes the implementation of the platform. Finally, Section 5 presents the conclusion of the paper.
The manufacturing platforms aim to satisfy the needs of the shop floor. This manufacturing platform can reduce production costs, detect anomalies, schedule production plans, or manage production processes in real-time [
Numerous kinds of research have been conducted to solve the problem of shop floor scheduling, also known as the job shop scheduling problem (JSSP). The JSSP is proved to be the NP-hard problem [
Job 1 | 1 | 31 | 0 | 24 | 2 | 50 | 3 | 86 | 4 | 72 |
Job 2 | 2 | 87 | 1 | 98 | 3 | 46 | 0 | 61 | 4 | 78 |
Job 3 | 4 | 43 | 3 | 75 | 2 | 20 | 1 | 32 | 0 | 53 |
This aspect of the JSSP has some limitations on the existing field systems [
Scheduling textile manufacturing has been studied in numerous studies, and the textile processes used are different. Saydam et al. [
The SMEs manufacturing platform has four operation modules: user interface, platform server, database, and job shop scheduling engine. The users of the platform are categorized as buyers and producers. In general, the buyers refer to companies that merchandise finalized products with a direct market connection—for example, the fashion companies that sell knit, dress, or other apparel. The producer is a manufacturer or vendor company that produces resources for producing finished products. In this sense, depending on the user type, the user interface of the platform serves two different purposes. First, the user interface gives a buyer with notification and monitoring services. The notification service provides buyers with scheduling results by utilizing the notification function of the platform. The process monitoring service allows buyers to track the production status of their orders in real time. Second, the user interface provides producers with notification, scheduling, and scheduled processing monitoring services. The notification service allows producers to notify buyers with real-time confirmed order results and production progress information. The scheduling service will connect producers to operate the job shop scheduling engine for scheduling execution. Finally, the scheduled processing monitoring service is a service that shows the history and progress of the production information.
The job shop scheduling engine operates as a module that handles scheduling-related tasks such as scheduling execution, scheduling history monitoring, and scheduling result or historical data storage. The structure of the proposed SMEs manufacturing platform is depicted in
All modules are linked to each other and have a high dependency. The primary functions of the platform are order management, product management, company management, facility management, scheduling, and evaluation. The overall operation model of the platform is as follows.
Step 1: Buyers register their orders on the platform through the order module. The buyers must also provide their expected delivery date for the scheduling process. After the buyers execute the order process, producers check the order lists to activate the scheduling system.
Step 2: Producers obtained the scheduling results from the scheduling module and estimated production delivery dates. The producers notify the buyers with a notification function.
Step 3: When more than one producer suggests the possibility of processing their orders, the buyers select the production company to progress the production.
Step 4: The producers confirm the confirmed orders and the schedule to start the production process.
In this paper, the weaving process data of SME-owned manufacturing are used to test the GA model. There are three main operations in textile manufacturing: weaving and knitting, dyeing, and sewing and post-processing. First, the weaving process produces simple fabrics for clothes or shoes. This simple fabric results from various patterns and designs that require a high level of technology. Depending on the type of weaving, the processing time may be longer than other designs for the finest and sheer fabric quality. Second, the dyeing process dyes the fabrics. This process requires diverse temperature settings that require more energy than other processes. Finally, the sewing and post-processing finalize the textile manufacturing. On the other hand, if the product the buyer wants to order is fabric, sewing and post-processing are not always required—only apparel goods are required to proceed with the process. Since the weaving process is longer than other processes, efficient scheduling is required. Generally, the weaving process takes more than three months on one machine. In some cases, it lasts more than six months. These practices of the weaving process can delay the production time of manufacturing. According to a condition of JSSP, the operation should be declared before executing the scheduling. However, as the study focuses on solving real-world problems, textile manufacturing does not always cover all three processes on one production site. Thus, parallel operation of the same job on different machines is allowed in this weaving process scheduling.
The proposed approach is tested by scheduling examples from the SME-owned weaving factory. The raw data were collected for one year, from January 1, 2020, to December 31, 2020. The total data are 62,452 records, and we used 11% of the total data in this study. There are three conditions in data selection. First, the data selection was based on machine numbers from 1 to 20. Second, among the selected machines, the jobs need to have lasted more than three days. Third, the chosen machines have produced two or more products on each machine. Utilizing the inputs for the GA model, the raw data are required to be preprocessed. There are seven columns in raw data: work date
Notations | Definition |
---|---|
Number of jobs | |
Number of machines | |
Processing time | |
Product list from raw data | |
Rate of absenteeism (set as 5%) | |
Density of |
|
Number of employees in manufacturing (set as 40) | |
rpm of |
|
Mean value of rpm | |
Processing time of |
|
Work start date of order |
|
Work due date for order |
There are three steps in raw data preprocessing. First, the number of products is divided based on the input data
Product name | Buyer | Producer | |||||
---|---|---|---|---|---|---|---|
Demand | Delivery |
Processing time/day | Available machine | Work start date | |||
6CL-FJ | 34,686 | 2020/08/01 | 49 | 555 | 27,360(19) | 2020/07/01 | |
6KOS-FD2 | 23,605 | 2020/07/11 | 67 | 560 | 86,400(60) | 2020/05/10 | |
6NF-1493-1 | 364,351 | 2020/09/06 | 76 | 554 | 168,480(117) | 2020/05/10 | |
ECO-SL157 | 220,455 | 2020/08/24 | 176 | 516 | 74,880(52) | 2020/07/01 | |
ES FD TSL R/S | 49,406 | 2020/08/26 | 60 | 509 | 77,760(54) | 2020/07/01 | |
FBR 0505 | 33,447 | 2020/07/10 | 183 | 507 | 8,640(6) | 2020/07/01 | |
FD 210T R/S REC | 136,268 | 2020/08/17 | 81 | 544 | 69,120(48) | 2020/05/10 |
The final step of the raw data preprocessing is to save a list of input data into the .csv file. After this process, the GA model is ready for the execution process. In the execution process of the GA model, setting up the three values is important. The three inputs are population size, iteration count, and mutation rate. The initiation of the model begins with setting up the population size to generate a set of chromosomes so that the first operation of the algorithm, selection, can be used. Iterations are initiated depending on the population size, and each iteration randomly generates a set of chromosomes within the range multiplied by
Then the producer can execute the GA model. According to the
In this paper, the data are divided into individual orders according to the average production cycle (three months/six months) of each product and 20 machines from the actual data. We test the feasibility of the GA model in textile manufacturing by comparing manual scheduling from field experts in the weaving process in textile manufacturing owned by SMEs, i.e., manual scheduling. We used Intel Core i-7 with 32 GB memory at a speed of 2.9 GHz and Python 3.9 to solve the JSSP and run the platform. Since the products are divided into multiple
Orders | GA (days) | Manual scheduling (days) | ||
---|---|---|---|---|
Order 1 | 40 | 20 | 152 | 155 |
Order 2 | 38 | 13 | 174 | 190 |
Order 3 | 27 | 8 | 136 | 135 |
Order 4 | 63 | 19 | 118 | 136 |
As shown in
The job shop scheduling execution is desired to get the optimized schedule for the production plan. Optimized schedules are based on the shortest makespan. Producers are expected to be able to run as many jobs as possible. When the scheduling engine generates the makespan, producers select the jobs that end within expected due dates. The job shop scheduling execution process requires several input data to be prepared for its algorithm. We used python language and Django Web application framework for scheduling and matching systems in this platform.
The JSSP algorithm in the platform is based on the GA model. Shao et al. [
Types | Articles | Optimization objectives | Algorithm |
---|---|---|---|
JSSP | [ |
Makespan, high feature extraction of hidden pattern during scheduling using auto-encoder | Self-supervised long-short term memory with GA |
FJSP | [ |
Makespan, distributed manufacturing environment | Hybrid GA algorithm |
DFJSP | [ |
Makespan, real-world data encoding | Refined GA algorithm |
Proposed system | Makespan, delivery date | Proposal of framework using GA algorithm |
In this paper, we present a novel method of manufacturing platform by enabling the scheduling process. The advantages of our system are to help producers to make faster and optimal decisions for the production schedule by executing the job shop scheduling engine and broadening the opportunity to attract buyers. Buyers can have a chance to discover new SMEs and obtain reasonable prices from the producers. Moreover, this order management, product information management, and dashboard menu allow users to track their orders and manage their production site conditions in real time. This paper uses the GA to solve the JSSP to obtain the optimal schedule. To test the feasibility of the JSSP model, we used textile manufacturing data, which is focused on the weaving process. The raw data are preprocessed for input setup by utilizing the weaving process to the GA model of JSSP. The procedure for executing the JSSP model is as follows. First, calculate the total processing time based on the features of the product. Second, separate jobs based on the number of machines and processing time. Third, make a job order table and processing timetable to operate the GA. Fourth, design and implement the system. The platform design aims to connect buyers and producers in one place to obtain the benefit of short delivery time, efficient scheduling, increased opportunities to discover new producers and buyers, and fast response. Finally, display the result with a Gantt chart. However, in the future, the automation of the job shop scheduling engine will predict the optimal rpm value that will affect processing time, as the current way of executing the job shop scheduling engine requires producers to input the rpm value for each product manually. Therefore, we will implement advanced algorithms using deep learning methods to automate the system and improve its applicability to real-world manufacturing factories.