Research Article
1596
Application Statistical Quality Control P-Chart for Short Production in 3D Printied
COVID-19 Mask Production
Didit Damur Rochman
1, Asep Anwar
2, Riki Ridwan Margana
3, Rendiyatna Ferdian
41Industrial Engineering Widyatama University Bandung, Indonesia 2Industrial Engineering Widyatama University Bandung, Indonesia 3Industrial Engineering Widyatama University Bandung, Indonesia 1diditdr@widyatama.ac.id,2 asep.anwar@widyatama.ac.id,
3riki.ridwan@widyatama.ac.id,4rendiyatna.ferdian@widyatama.ac.id
Article History: Received: 10 January 2021; Revised: 12 February 2021; Accepted: 27 March 2021; Published
online: 20 April 2021
Abstract: The increasing transmission of the corona virus (covid 19) in the world has caused tremendous impacts on various things such as health, the economy and various other aspects of life. One of the causes of the increase in the spread of this virus is due to the lack of understanding of the public in the use of good masks. Many people use cloth masks that can be washed several times, but have a weakness that the face area is not tightly closed, resulting in leaks both in the nose area near the eyes and on the chin. We, from a team from the production systems laboratory of the Industrial Engineering Departement Widyatama University, provide solutions for the community by making masks that are produced using 3D Print. The production of COVID-19 masks using a filament fused 3D printer was carried out in September 2020 using PETG material because this plastic is food grade and does not require special treatment such as a heated chamber. To control the quality of the production process in making this 3D mask, a standardized P-chart is used. At the beginning of production there is an eratic but controllable failure rate so that it decreases in subsequent periods. Standardized P-charts can be used for small medium busines without the use of special computing tools and applications.
1. Introduction
The novel CoronaVirus-19 was declared by WHO) to become a global pandemic on March 11, 2020 (WH0, 2020, and until January 2021 it infected 84 million in the world and 751 thousand in Indonesia (Dong et.al, 2020). This condition has made the economy sink. on the verge of a recession if it is not controlled by both the Indonesian government's health department and the active role of the Indonesian people. Despite following directions from WHO to the public not to travel to high-risk areas, contact with symptomatic individuals, wash hands frequently and use face masks ( WHO, 2020b) there is still an increase in infections in Indonesia. Lack of discipline and disinformation is one of the root causes of handling the COVID-19 pandemic in Indonesia is not optimal. The results of a survey by the Central Bureau of Statistics (BPS) stated that 92% of people use masks but their behavior is to keep their distance in social interaction and hand washing is quite low (KPCPEN, 2020). Masks are common yes, used by the community is a cloth mask that can be washed many times, but it has a weakness that the face area is not tightly closed, resulting in leaks both in the nose area near the eyes and on the chin. The CDC recommends using nose wires on cloth masks to increase the filtration of respiratory air (CDC, 2020), but cloth masks that are widely used by the public do not apply these nose wires.
The solution proposed by the team from the production system laboratory of the Industrial Engineering Departement, Widyatama University for the part of the community with a high intensity of social interaction during this pandemic is to choose a face mask design that will be produced with 3D Print (Rochman et.al, 2020). The production of facial masks from the selected design uses 2 3D printers with filament fused in the Production System laboratory, each with a bed size of 200mmx200mm and 300x600mm. The proceeds from the production of this face mask will be donated to the community in the village 7km from the university as many as 450 pieces as part of community service carried out by academics.
The flexibility of the 3D printer in producing items for prototype scale before mass production using conventional machining is very high, but to carry out production in large enough quantities, product quality problems arise, namely the level of defects that is quite high both because of problems with the material and due to machine reliability. This paper will discuss the use of the P control chart for short production run face masks to reduce covid-19 infection.
Research Article
1597
The production of COVID-19 masks using a filament fused 3D printer was carried out in September 2020 using PETG material because this plastic is food grade and does not require special treatment such as a heated chamber. This mask printing process is as shown in Figure 1.In the first step, STL loading was carried out on the SLIC3R application consisting of 3 parts, namely the gasket, cap and body as shown in Figure 2.
Furthermore, optimization of the number of parts is carried out based on the size of the bed of the 3D printer in the lab, namely 200mmx200mm (small) and 300mmx600mm (large), which are the same 6 parts for large printers and 3 parts for small sizes. Small printers use gantry style like Prrusa I3 and large printers use COREXY so. After slicing STL into GCODE, 3 GCODE files were obtained for small printers and 3 GCODE for large printers. The parameters in the slicing process are as follows:
• Nozzle 0.5mm • Layer height 0.3mm • Speed 250mm/s • Travel Speed 350mm/s • Infill 20% • Filament Diameter 1.75mm • Nozzle Temperature 230C
Load and Optimize Number Part based bed size
Slicing STL to GCODE
Send GCODE to 3D Printer
Product Inspection
Short Production Run P-Chart
Improvement
Research Article
1598
a. Gasket b. Cap c. Body
Figure 2. Parts Masker
The estimated printing time for each part is in table 1 with the production of 9 face masks per batch. The number of face masks to be printed is 450 which will spend 1035 machine hours or 33 days on 2 shifts.
Tabel 1. Estimated Printing Time
Part Printer 200mmx200mm Printer 300mmx600mm
Gasket 30 minutes 30 minutes
CAP 35 minutes 40 minutes
Body 55 minutes 1.3 hours
The final product inspection of the 3D print results is carried out using the following criteria: • Dimension Stability compared to design.
• Surface finish • Failed/Success Print • Fitting Assembly
The results of the inspection are entered into MS Excel to control production quality using the Standardized P-Chart short production run control chart. Standardized P-P-Chart for short production run (Lai et.al, 1996) is used because the production cycle is very short such as in make to order production so there is not enough data to construct control charts (Montgomery, 2007). The constructs of the standardized P-Chart are:
Let 𝑋𝑘 = 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑛𝑜𝑛𝑐𝑜𝑛𝑓𝑜𝑟𝑚𝑖𝑛𝑔 𝑖𝑡𝑒𝑚 𝑖𝑛 𝑘𝑡ℎ 𝑛 𝑠𝑎𝑚𝑝𝑙𝑒 𝑝𝑘 =
𝑋𝑘
𝑛𝑘 fraction of non conforming item Standardized Pk is Zk 𝑍𝑘 = √ 𝑘 𝑘 − 1 √𝑛 (𝑝𝑘− 𝑝̅ − (1.2 𝑛⁄ )) √𝑝𝑘(𝑝𝑘− 𝑝̅) With lower control limit -3 and upper control limit +3
Production split in 2 batch with 16 days per batch and produce 225 face mask per batch, everyday produce 15 face mask full assembled.
3. Result and discussion
During production, quality inspections are carried out on each result of 3D printing and product assembly. The defective product criteria that have been defined, such as dimensional stability and fitting assembly, are the most dominant types of failure besides printing failure.
Table 2. Number of Reject Item
Dimensional stability and print failure make a large contribution to defective products. After elaboration, this failure generally occurs due to the humid factor of the PETG filament so that air bubbles occur in the extrusion process and the layer cohesion decreases. The process of enhancement is carried out by replacing with a new filament and drying the old filament by inserting it into a container filled with silica gel.
Using the equation based on Lai et.al (1994), the calculation of the proportion of defects, the Zk value for production batches 1 and 2 is calculated and displayed in table 1.Zk calculations can be seen that there is a correction factor depending on the value of k, the greater the value of k, the correction factor will be getting smaller. The results of the Zk calculation are as shown in Table 3, and are plotted in Figure
Criteria Count Percent
Dimension Stability compared to design. 17 40,48%
Failed Print 15 35,71%
Fitting Assembly 2 4,76%
Research Article
1599
Table 3. Standardized P- Chart
Period 1 to 8 in the production of the 1st batch can be seen that the product failure rate appears eratic and starts to stabilize from period 9 onwards. This happens because the 3D printer tuning is still happening to produce a better product by changing printer parameters. Another factor that emerged was the existence of several hours of electrical outage in the 5th period which resulted in a total of 4 pieces of printing failure. Improvement is done by adding a UPS to the 3D printer so that if there is an electricity outage it can still print for 30 minutes.In the next period, the defect rate can be better controlled, but there are still defects that arise due to wear of the 3D printer belt but no replacement is made because the leadtime for ordering spare parts reaches 3 weeks.
The calculation results in table 1 are then plotted on the P-Chart chart with UCL and LCL constant at +3 and 3, shown in figure 3 and 4.
Figure 3. Standarized P-Chart Batch 1
Batch Sample Number Sample Size Defect pk Zk
1 1 15 4 0.2667 0.7222 2 15 2 0.1333 -1.1869 3 15 1 0.0667 -1.9841 4 15 3 0.2000 -0.0676 5 15 4 0.2667 0.8074 6 15 2 0.1333 -0.9194 7 15 3 0.2000 -0.0632 8 15 1 0.0667 -1.7318 9 15 2 0.1333 -0.8902 10 15 1 0.0667 -1.7076 11 15 1 0.0667 -1.6991 12 15 1 0.0667 -1.6920 13 15 2 0.1333 -0.8735 14 15 1 0.0667 -1.6811 15 15 1 0.0667 -1.6769 16 15 1 0.0667 -1.6731 2 1 15 2 0.1333 -0.8393 2 15 1 0.0667 -2.2910 3 15 2 0.1333 -1.0279 4 15 1 0.0667 -1.8706 5 15 2 0.1333 -0.9383 6 15 1 0.0667 -1.7746 7 15 1 0.0667 -1.7498 8 15 1 0.0667 -1.7318 9 15 1 0.0667 -1.7183 10 15 3 0.2000 -0.0617 11 15 1 0.0667 -1.6991 12 15 1 0.0667 -1.6920 13 15 1 0.0667 -1.6861 14 15 1 0.0667 -1.6811
Research Article
1600
Figures 3 and 4 show statistically controlled processes, because all plot points are on the control chart control boundaries. The trend of the proportion of defects in both the 1st and 2nd batches can be seen to be decreasing with the dominance below the center line of the control chart.Figure 4. Standarized P-Chart Batch 2
4. Conclusion
From this study it can be concluded that the standardized P-chart can be used to control the quality of the face mask production process using a 3D printer. At the beginning of production there is an eratic but controllable failure rate so that it decreases in subsequent periods. Standardized P-charts can be used for small medium busines without the use of special computing tools and applications.
In further research, it is necessary to address the effect of learning effects on production and its disabilities by using a short run control chart. The learning curve will provide an overview of the relationship between the machine operator's level of mastery and the level of production defects.
5. References
1. World Health Organization. (2020). WHO Director-General's opening remarks at the media briefing on COVID-19-11 March 2020
2. Dong, E., Du, H., & Gardner, L. (2020). An interactive web-based dashboard to track COVID-19 in real time. The Lancet infectious diseases, 20(5), 533-534.
3. World Health Organization (2020b). Novel Coronavirus (2019-nCoV) Advice for the Public, https://www.who.int/emergencies/diseases/novel-coronavirus-2019/advice-for-public
4. Komiter Penanganan COvid-19 dan Pemulihan Ekonomi Nasional (2020), Hasil Survei BPS: 92 Persen Warga Patuh Pakai Masker Selama Pandemi COVID-19, https://covid19.go.id/p/berita/hasil-survei-bps-92-persen-warga-patuh-pakai-masker-selama-pandemi-covid-19
5. CDC (2020), Improve the Fit and Filtration of Your Mask to Reduce the Spread of COVID-19, https://www.cdc.gov/coronavirus/2019-ncov/prevent-getting-sick/mask-fit-and-filtration.html 6. Didit Damur Rochman, Asep Anwar, Riki Margana, Rendiyatna (2020), 3D PRINT COVID-19
MASK DESIGN SELECTION USING ANALYTICAL HIERARCHY PROCESS,
https://repository.widyatama.ac.id/xmlui/handle/123456789/12159
7. Jabarullah, N. H., Surendar, A., Arun, M., Siddiqi, A. F., & Krasnopevtseva, T. O. (2020). Microstructural Characterization and Unified Reliability Assessment of Aged Solder Joints in a PV Module. IEEE Transactions on Components, Packaging and Manufacturing Technology, 10(6), 1028-1034.
8. Montgomery, D. C. (2007). Introduction to statistical quality control. John Wiley & Sons
9. Lai K. Chan Brian D. Macpherson Peter H. Xiao, (1996),"Standardized p control charts for short runs", International Journalof Quality & Reliability Management, Vol. 13 Iss 6 pp. 88 - 95
