Process design for the management of changes in manufacturing: Toward a Manufacturing Change Management process

Koch, Jonas; Gritsch, Achim; Reinhart, Gunther

doi:10.1016/j.cirpj.2016.04.010

Cited by 29 publications

(8 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Not only the methods from the abovementioned literature studies but also a wide range of machine vision and image processing algorithms are available for developers to utilize [ 3 ]. Ideally, an almost perfect AOI system can be developed if all the defect types are reported and studied in advance.…”

Section: Introductionmentioning

confidence: 99%

Printed Circuit Board Defect Detection Using Deep Learning via A Skip-Connected Convolutional Autoencoder

Kim

Choi

et al. 2021

Sensors

103

View full text Add to dashboard Cite

As technology evolves, more components are integrated into printed circuit boards (PCBs) and the PCB layout increases. Because small defects on signal trace can cause significant damage to the system, PCB surface inspection is one of the most important quality control processes. Owing to the limitations of manual inspection, significant efforts have been made to automate the inspection by utilizing high resolution CCD or CMOS sensors. Despite the advanced sensor technology, setting the pass/fail criteria based on small failure samples has always been challenging in traditional machine vision approaches. To overcome these problems, we propose an advanced PCB inspection system based on a skip-connected convolutional autoencoder. The deep autoencoder model was trained to decode the original non-defect images from the defect images. The decoded images were then compared with the input image to identify the defect location. To overcome the small and imbalanced dataset in the early manufacturing stage, we applied appropriate image augmentation to improve the model training performance. The experimental results reveal that a simple unsupervised autoencoder model delivers promising performance, with a detection rate of up to 98% and a false pass rate below 1.7% for the test data, containing 3900 defect and non-defect images.

show abstract

Section: Introductionmentioning

confidence: 99%

Printed Circuit Board Defect Detection Using Deep Learning via A Skip-Connected Convolutional Autoencoder

Kim

Choi

et al. 2021

Sensors

103

View full text Add to dashboard Cite

show abstract

“…In change management (CM), throughout the entire product lifecycle focus is on the task of modifying a released system (Eisa, Garstenauer, and Blackburn 2018;Karthik and Reddy 2016). These systems could be products (engineering changes (EC); Ullah, Tang, and Yin 2016), or factory systems (manufacturing changes (MC); Koch, Gritsch, and Reinhart 2016). Changes in these systems could be either triggered by problems (emergent changes) or by the introduction of new product and process specifications (initiated changes; Eckert, Clarkson, and Zanker 2004).…”

Section: Challenges and Characteristics That Influence Change Propagamentioning

confidence: 99%

“…It is also evident from the literature review that there is a reported gap that assessment of ECs lacks in the consideration of other domains, especially of the process domain (Siddharth and Sarkar, 2017;Ullah et al, 2017;Plehn et al, 2016;Koch et al, 2016;Hamraz and Clarkson, 2015;Hamraz et al, 2012;Koh et al, 2012).…”

Section: Literature Gap and Its Relevancementioning

confidence: 99%

Characteristics of changeable systems across value chains

Masood

Kern

Clarkson

2020

International Journal of Production Research

View full text Add to dashboard Cite

Engineering changes (ECs) are inevitable for businesses due to increasing innovation, shorter lifecycles, technology and process improvements and cost reduction initiatives. The ECs could propagate and cause further changes due to existing system dependencies, which can be challenging. Hence, change management (CM) is a relevant discipline, which aims to reduce the impact of changes. EC assessment methods form the basis of CM that support in assessing system dependencies and the impact of changes. However, understanding of which factors influence the changeability across value chains (VCs) is limited. This research adopted a VC approach to EC assessment. Dependencies in products and processes were captured, followed by risk (i.e. likelihood x impact) assessment of ECs using change prediction method (CPM). Four industrial case studies were conducted (3x automotive, 1x furniture manufacturing) to identify design (product) and manufacturing (process) elements with high risk to be affected by ECs. Based on the case results, characteristics were identified that influence changeability across VC. This contributed to the CM domain while businesses could also use the results to assess ECs across VC, and improve the design of products and processes by increasing their changeability across VC e.g. by proactive decoupling or reactive handling of system dependencies.

show abstract

“…In the field of Manufacturing Change Management, Koch developed an MCM context model [4] and an MCM reference process [22], which enable scientists to locate their research within MCM. Since the context model's objective is to provide a framework, its level of detail is too low for the identification of MCs.…”

Section: State Of Research Regarding the Identification Of Alternativmentioning

confidence: 99%

Methodology for the identification of alternative manufacturing changes for safety–critical components

Bergs

Hermann

Rey

et al. 2020

Prod. Eng. Res. Devel.

View full text Add to dashboard Cite

A variety of factors are forcing companies to change their products and the components these products are composed of. Examples of current factors are changing customer requirements, product optimizations or ever-shorter product lifecycles. These component changes, particularly in safety-critical components in the medical and aerospace industry, pose great challenges and are limited in scope. The challenges and limitations arise from the high costs and the amount of time involved in component recertification as required by law, as well as revalidations of the manufacturing after a change has been implemented. Some component changes are even mandatory. In this case there are no options for reducing recertification efforts by choosing between different design changes. In contrast, there are alternative manufacturing options with which to implement the prescribed component changes that can reduce the amount of revalidation required on the manufacturing side. Due to the fact that changes to the manufacture of safety-critical components require validation, the existence of alternative manufacturing changes offers the potential to save time and money by selecting the best-suited alternative. In order to be able to select the best-suited alternative, all alternative manufacturing changes must be known, which poses a challenge for technology planners dealing with the complex manufacture of safety-critical components. Therefore, a methodology is introduced to support technology planners in identify alternative manufacturing changes with which to implement scope-restricted changes to safety-critical components. The methodology is validated by applying it to a case study from the medical industry.

show abstract

Process design for the management of changes in manufacturing: Toward a Manufacturing Change Management process

Cited by 29 publications

References 27 publications

Printed Circuit Board Defect Detection Using Deep Learning via A Skip-Connected Convolutional Autoencoder

Printed Circuit Board Defect Detection Using Deep Learning via A Skip-Connected Convolutional Autoencoder

Characteristics of changeable systems across value chains

Methodology for the identification of alternative manufacturing changes for safety–critical components

Contact Info

Product

Resources

About