INDUSTRY 4.0: INTELLIGEN
T QUALITY
CONTROL AN
D SURFACE DEFECT
DETECTION
Vineeth C Johnson
KLE Technological University, Hubballi-31, Karnataka, (India).
E-mail: vineethcjohnson@gmail.com
Jyoti S Bali
KLE Technological University, Hubballi-31, Karnataka, (India).
C B Kolanur
KLE Technological University, Hubballi-31, Karnataka, (India).
Shilpa Tanwashi
KLE Technological University, Hubballi-31, Karnataka, (India).
Reception: 02/11/2022 Acceptance: 17/11/2022 Publication: 29/12/2022
Suggested citation:
Johnson, V. C., Bali, J. S., Kolanur, C. B., y Tanwashi, S. (2022). Industry 4.0: Intelligent Quality Control and
Surface Defect Detection. 3C Empresa. Investigación y pensamiento crítico, 11(2), 186-192. https://doi.org/
10.17993/3cemp.2022.110250.186-192
https://doi.org/10.17993/3cemp.2022.110250.214-220
3C Empresa. Investigación y pensamiento crítico. ISSN: 2254-3376
Ed. 50 Vol. 11 N.º 2 August - December 2022
214
ABSTRACT
Quality Control (QC) has recently emerged as a significant global trend among manufacturers,
adopting intelligent manufacturing practices in view of Industry 4.0 requirements. Intelligent
manufacturing is the process of enhancing production through the use of cutting-edge technologies,
sensor integration, analytics, and the Internet of Things (IoT). The proposed paper mainly focuses on
the study of the scope and the evolution of quality control techniques from conventional practices to
intelligent approaches along with the state of art technologies in place. The challenges faced in
building intelligent QC systems, in terms of security, system integration, Interoperability, and Human-
robot collaboration, are highlighted. Surface defect detection has evolved as a critical QC application
in modern manufacturing setups to ensure high-quality products with high market demand. Further,
the recent trends and issues involved in surface defect detection using intelligent QC techniques are
discussed. The methodology of implementing surface defect detection on cement wall surfaces using
the Haar Cascade Classifier is discussed.
KEYWORDS
Quality Control, Industry 4.0, Internet of Things, Intelligent manufacturing, Interoperability, cutting-
edge technologies, analytics, surface defect detection.
1. INTRODUCTION
Industrial societies are increasingly interested in intelligent manufacturing, especially with the advent
of Industry 4.0, which calls for the majority of industrial tasks to be performed by robots with
intelligence. It expressly indicates that the production systems will be fully connected and all
production processes, including quality control and administration, can be made as intelligent as
possible to run with the least amount of human involvement. Interoperability is a well-known
necessity for the quick transformation of industry-specific processes. Hence it calls for integrating
quality functions with other manufacturing operations to maintain intelligent collaboration so that
quality-related knowledge may be shared with other manufacturing processes. On the other hand, the
integration of manufacturing processes has ensured better performance.
Quality Control (QC) refers to a policy or set of practices created to satisfy a client's or customer's
requirements or to fulfil a defined group of quality standards for a manufactured product or service
[1]. It plays a significant role in maintaining and improving the quality of manufactured products. It
involves testing the products to determine that they meet the necessary specifications. Testing is done
to determine whether corrective measures are needed in finetuning the manufacturing processes to
meet customer demands. QC ensures additional benefits such as reduced inspection and production
costs, minimization of variations, and cost-effective use of resources. The process inspires employees
to create high-quality goods leading to greater customer satisfaction [2]. Establishing customer-
acceptable quality standards, finding defects or variations in the raw materials and manufacturing
processes, ensuring smooth and uninterrupted production, assessing the degree of quality deviation in
a product during the manufacturing process, thoroughly examining the contributing factors and thus
achieving the objectives of quality control [3]. Some of the applications in that quality control are
involved include preserving the quality of processes, products, and services, alerting for process
abnormalities and fault detection, predicting the behaviour of machines, devices, and respective
equipment in terms of the expected yield, machine maintenance and condition monitoring. Thus these
practices ensure the effectiveness of the entire supply chain, starting from suppliers to customers [4].
The stepwise processes involved in quality control, starting from the inspection of manufactured
products to meet the specified requirements up to the decision of acceptance/ rejection, are shown in
Figure 1.
https://doi.org/10.17993/3cemp.2022.110250.214-220
215
3C Empresa. Investigación y pensamiento crítico. ISSN: 2254-3376
Ed. 50 Vol. 11 N.º 2 August - December 2022
Figure 1. Quality control flowchart.
2. STATE-OF-THE-ART TECHNOLOGIES FOR QUALITY
CONTROL
Production development is still guided by scientific management theory and has evolved from manually
operated systems in Industry1.0 (I1.0) to intelligent systems in Industry4.0(I4.0). This strategy starts the
management concept's scientific and technological growth. Control charts use graphical representations to
determine whether the manufacturing processes or final products meet the intended specifications.
Economic statistics defines econometrics as a field that helps to monitor and promote the economic factors
influencing the quality of products. The Taguchi methods determine the ideal cost of quality throughout a
product's life and emphasize product design and development in reducing defects in the manufactured
products.
Quality costs involve a model that tracks the quality costs and is used to keep tabs on quality management's
outcomes. The systems theory is used to model corporate systems through hierarchical breakdown and
feedback. The primary organizational innovation paradigm for quality improvement in Japanese
corporations was Quality circles, a collection of goals representing the best worldwide practices as a
benchmarking model for building products and processes. A "flat" organizational structure with a specific
position for quality management is the foundation of the lean system, an organizational, technical system
then established in Japan. The material, process, and information flow from raw materials to goods
delivered to customers are optimized and synchronized by the Quality Management supply chain model,
which optimizes inventory and lowers the costs associated with product life. The concept of digital quality
for intelligent systems has been improved by virtual quality using simulation models for the intelligent
model quality of the integrated management system [5].
3. QUALITY CONTROL SYSTEMS WITH INTELLIGENCE
Industry 4.0, which calls for most industrial tasks to be performed by robots with intelligence, has
made intelligent manufacturing popular with industrial societies. Setting up knowledge-intensive tasks
to ensure quality and continual improvement is the fundamental prerequisite for developing
intelligence in quality control. This calls for systematic performance monitoring and evaluation.
Computer vision technology is mainly used to replace human eyes because many manufactured
products' dimensions and surface characteristics dictate their quality. It is well known that automated
machine vision systems can evaluate geometric and surface features to judge product quality and
apply statistical analysis methods. Artificial intelligence can be used in industrial system design to
https://doi.org/10.17993/3cemp.2022.110250.214-220
3C Empresa. Investigación y pensamiento crítico. ISSN: 2254-3376
Ed. 50 Vol. 11 N.º 2 August - December 2022
216
ensure such capabilities. The manufacturers may require the use of lean manufacturing techniques and
sophisticated machinery. To maintain proactive management of equipment, processes, services, and
goods, they must increase predicted output while increasing the productivity and efficiency of their
manufacturing hardware. Intelligent machinery should be able to process massive data collected from
various sensors and use artificial intelligence and machine learning techniques. Utilizing cutting-edge
manufacturing technologies is necessary to create intelligent systems that can be reconfigurable,
interoperable, and reusable and cut down on potential wastes like scraps, overtime, and expenditures.
AI-enhanced sensors, big data analytics-based decision support systems, and improved materials
should be integrated throughout the industrial life cycle and serve as its drivers. A manufacturing
system can predict and comprehend critical events and solve problems instantly before they result in
any hazardous and dangerous situations or wastes, given the ability to process real-time data collected
from the machines and conduct intelligent analysis over those through AI technologies. Thus it allows
manufacturing systems to perform preventative maintenance and create fully functional manufacturing
suits [4].
To provide the resources and capabilities required for satisfying standards and continuously enhancing
the efficiency of the quality system, intelligent quality requires management commitment. The
objective of the design of the manufacturing chain should be to maintain a constant and sustained level
of quality for operations and services across the entire organization, with a high degree of integration.
Another part of intelligent quality focuses on utilizing a knowledge-driven strategy rather than a
conventional data-driven one. The system needs the appropriate operational and quality knowledge to
establish the necessary level of intelligence. Methodologies and techniques are required to extract
knowledge from the data that is currently accessible to produce self-behaviour with the desired quality
function, as shown in Figure 2 [4].
Figure 2. Robot manipulators for intelligent quality control.
Source: https://www.therobotreport.com/top-5-countries-using-industrial-robots-2018/.
4. CHALLENGES IN INTELLIGENT QUALITY CONTROL
❖Security Issues in Smart Manufacturing
An intelligent manufacturing system uses an integrated network to share information between
manufacturing or machining units and end users using the internet. A globally unique identity and end-
to-end data encryption are required for internet-based information sharing to ensure data and
information security throughout the system. Therefore, every network node must be secured against
outside threats and data exploitation [6].
❖System Integration
https://doi.org/10.17993/3cemp.2022.110250.214-220
217
3C Empresa. Investigación y pensamiento crítico. ISSN: 2254-3376
Ed. 50 Vol. 11 N.º 2 August - December 2022
Integrating new and existing technology equipment is a hurdle in developing an intelligent
manufacturing system. A better communication system is also necessary for machine-to-machine
communication and system interconnectivity. IPv6 connectivity is needed for the most modern
production systems to enable more devices to be linked concurrently [6].
❖Interoperability
The capacity of several systems to comprehend and utilize one another's features independently using
proper synchronization of the communication protocols and standards is known as Interoperability.
The differences in transmission bandwidth, communication method, operational frequency, hardware
capabilities, etc., limit the system's compatibility.
❖Safety in Human-Robot Collaboration
Any instructions given in a human language should be converted into machine language by the
multilingual intelligent manufacturing systems for performing the appropriate action. The instructions
can be in verbal or text form of input from the operator.
❖Multilingualism
Financial Analysis and return on investment (ROI) should be thoroughly examined for an existing
manufacturing system before transitioning to other advanced technologies. The increased expenditure
necessary compared to production losses during an upgrade and the time needed to recover the
investment's return [6].
❖Surface defect detection
The large and complicated curved-surface components present a challenge for inspection since it is
challenging to assure shape accuracy using conventional inspection techniques, which is crucial to
these components' functional performance [7]. The manual inspection methodology's accuracy for
inspecting these components falls short of expectations. Thus it calls for the requirement of an
inspection system that uses mobile manipulators to measure huge components with complicated
curved surfaces in three dimensions accurately and automatically [8]. Surface defects are the lines or
planes that divide a substance into sections, each having a distinct orientation but the same crystalline
structure. Surface defects are often caused by surface finishing techniques like embossing, weather-
related degradation, or environmental stress cracking [9]. Defects may also be produced when metals
are used and treated for industrial reasons.
The aircraft and automotive industries reject any material with manufacturing defects since even a tiny
defect in a finished product could result in a catastrophe. Steel rolling processes cannot do real-time
inline surface flaw checking without automatic machine vision technology. Some spots may go
unnoticed, costing production time and causing significant financial losses. Modern methods for
finding surface defects include eddy current testing, electrical resistance testing, flux leakage testing,
magnetic testing, thermographic testing, radiographic testing, resonant testing, ultrasonic testing,
penetrant testing, and visual testing. A successful surface defect detection would ultimately identify all
the surface defects on the product's surface, eliminate products with defective surfaces, and help repair
the faulty surfaces. Automated visual inspection techniques for surface defect detection can help
intelligent quality control automatically inspect each printed circuit board assembly and update
measurements and observations without human interaction [10-13].
5. METHODOLOGY OF IMPLEMENTATION OF SURFACE
DEFECT DETECTION
Here the process of implementation of intelligent Surface defect detection is presented further. Due to
its high inference speed, the object detection model was trained using the Haar Cascade algorithm.
This algorithm uses a set of 1000 images for training and uses Haar features that traverse through the
entire image to detect a location of defects, namely, spalling, cracks, uneven surfaces, and holes. The
samples of test images for defect detection are shown in figure 3.
https://doi.org/10.17993/3cemp.2022.110250.214-220
3C Empresa. Investigación y pensamiento crítico. ISSN: 2254-3376
Ed. 50 Vol. 11 N.º 2 August - December 2022
218
The experimentation results show that the Haar Cascade classifier is superior to other classifiers based
on deep learning models in terms of higher inference speed. But the classifier suffers from some
drawbacks, namely, low accuracy, false positives detection, and difficulty in training using custom
datasets. The performance of the Haar Cascade classifier is evaluated in terms of inference time and
accuracy of classification. As attaining accuracy in surface defect detection is crucial, autonomous
quality inspection systems must adopt deep learning algorithms, namely, YOLO, Fast R-CNN, Faster
R-CNN, etc., for better accuracy.
6. CONCLUSIONS
The paper focuses on the study of quality control techniques, the evolution and processes involved and
the state of art technologies in use. A review of challenges in intelligent quality control is explicitly
done on surface defect detection. Further, the methodology and results of implementing surface defect
detection using the Haar Cascade classifier on cement walls are presented. The performance of the
Haar Cascade classifier is evaluated in terms of inference time and accuracy of classification. Further,
it is proposed to analyze the surface defect detection performance using AI-based classifiers.
REFERENCES
1. Agolla, J. E. 2020. Smart Manufacturing: Quality Control Perspectives. In P. Li, P. A. R. Pereira, &
H. Navas (Eds.), Quality Control - Intelligent Manufacturing, Robust Design and Charts.
IntechOpen. https://doi.org/10.5772/intechopen.95143.
2. Hayes, A. 2021, Understanding Quality Control. Investopedia; www.investopedia.com. https://
www.investopedia.com/terms/q/quality-control.asp
3. Sinha, D. 2014, Quality Control (QC): Definition, Importance and Tools of Quality Control. Your
Article Library; https://www.yourarticlelibrary.com/production-management/quality-control-qc-
definition-importance-and-tools-of-quality-control/41085
Spalling Detected Crack detected
Residual surface detected Hole detected
Figure 3. Surface defects detected.
https://doi.org/10.17993/3cemp.2022.110250.214-220
219
3C Empresa. Investigación y pensamiento crítico. ISSN: 2254-3376
Ed. 50 Vol. 11 N.º 2 August - December 2022
4. Oztemel, E. 2020. Introduction to Intelligent Quality Management. In P. Li, P. A. R. Pereira, & H.
Navas (Eds.), Quality Control - Intelligent Manufacturing, Robust Design and Charts.
IntechOpen. https://doi.org/10.5772/intechopen.94971
5. Majstorović
, V. D. 2011. QUALITY MANAGEMENT – STATE OF THE ART AND FUTURE
DEVELOPMENT. ResearchGate. https://www.researchgate.net/publication/
264652441_QUALITY_MANAGEMENT_STATE_OF_THE_ART_AND_FUTURE_DEVEL
OPMENT
6. Sudip Phuyal, Diwakar Bista, Rabindra Bista, 2020, Challenges, Opportunities and Future
Directions of Smart Manufacturing: A State of Art Review, Sustainable Futures, Volume 2,
100023, ISSN 2666-1888, https://doi.org/10.1016/j.sftr..100023.(https://
www.sciencedirect.com/science/article/pii/S2666188820300162)
7. Menderes Kam, Musa Ş
eremet.2021, Experimental and Statistical Investigation of Surface
Roughness and Vibration During Finish Turning of Aisi 4140 Steel workpiece under Cooling
Method, Surface Review and Letters, 28(10),2150092, https://doi.org/10.1142/
S0218625X2150092X
8. Hands, G., & Armitt, T. (n.d.). Surface and internal defect detection | SpringerLink. Surface and
Internal Defect Detection | SpringerLink; https://link.springer.com/chapter/
10.1007/978-94-011-4924-2_5?noAccess=true
9. Kam M, Demirtaş
M. 2021, Analysis of tool vibration and surface roughness during turning process
of tempered steel samples using Taguchi method. Proceedings of the Institution of Mechanical
Engineers, Part E: Journal of Process Mechanical Engineering.;235(5):1429-1438. https://
doi.org/10.1177/09544089211001976
10. What are Surface Defects? - Definition from Corrosionpedia. (n.d.). Corrosionpedia;
www.corrosionpedia.com.,https://www.corrosionpedia.com/definition/1614/surface-
defects#:~:text=Surface%20defects%20are%20usually%20formed,use%20of%20metals%20in
%20service
11. Chaudhary, Vikas & Dave, Ishan & Upla, Kishor. (2017). Automatic visual inspection of printed
circuit boards for defect detection and classification. 732-737. https://doi.org/10.1109/
WiSPNET.2017.8299858.
12. Kam M, İ
pekçi A, Argun K.2022, Experimental investigation and optimization of machining
parameters of deep cryogenically treated and tempered steels in electrical discharge machining
process. Proceedings of the Institution of Mechanical Engineers, Part E: Journal of Process
Mechanical Engineering.236(5):1927-1935. https://doi.org/10.1177/09544089221078133
13. Hongbin Jia, Y. L. Murphey, Jinajun Shi and Tzyy-Shuh Chang, 2004. An intelligent real-time
vision system for surface defect detection, Proceedings of the 17th International Conference on
Pattern Recognition, ICPR 2004. pp. 239-242 Vol.3, https://doi.org/10.1109/
ICPR.2004.1334512
https://doi.org/10.17993/3cemp.2022.110250.214-220
3C Empresa. Investigación y pensamiento crítico. ISSN: 2254-3376
Ed. 50 Vol. 11 N.º 2 August - December 2022
220
