Towards low-cost machine learning solutions for manufacturing SMEs

Kaiser, Jan; Terrazas, Germán; McFarlane, Duncan; Silva, Lavindra de

doi:10.1007/s00146-021-01332-8

Cited by 9 publications

(4 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition to the challenges in their own companies, various studies have also identified barriers on the market side (8). The available AI solutions that explicitly focus on the requirements and initial situations of SMEs are very limited (Kaiser et al 2021;Velmurugan et al 2021). Most applications are aimed at large companies and are tailored to their structures.…”

Section: Challenges Regarding Ai Implementation In Smesmentioning

confidence: 99%

Investigation of artificial intelligence in SMEs: a systematic review of the state of the art and the main implementation challenges

Oldemeyer,

Jede,

Teuteberg

2024

Manag Rev Q

View full text Add to dashboard Cite

While the topic of artificial intelligence (AI) in multinational enterprises has been receiving attention for some time, small and medium enterprises (SMEs) have recently begun to recognize the potential of this new technology. However, the focus of previous research and AI applications has therefore mostly been on large enterprises. This poses a particular issue, as the vastly different starting conditions of various company sizes, such as data availability, play a central role in the context of AI. For this reason, our systematic literature review, based on the PRISMA protocol, consolidates the state of the art of AI with an explicit focus on SMEs and highlights the perceived challenges regarding implementation in this company size. This allowed us to identify various business activities that have been scarcely considered. Simultaneously, it led to the discovery of a total of 27 different challenges perceived by SMEs in the adoption of AI. This enables SMEs to apply the identified challenges to their own AI projects in advance, preventing the oversight of any potential obstacles or risks. The lack of knowledge, costs, and inadequate infrastructure are perceived as the most common barriers to implementation, addressing social, economic, and technological aspects in particular. This illustrates the need for a wide range of support for SMEs regarding an AI introduction, which covers various subject areas, like funding and advice, and differentiates between company sizes.

show abstract

Section: Challenges Regarding Ai Implementation In Smesmentioning

confidence: 99%

Investigation of artificial intelligence in SMEs: a systematic review of the state of the art and the main implementation challenges

Oldemeyer,

Jede,

Teuteberg

2024

Manag Rev Q

View full text Add to dashboard Cite

show abstract

“…Each solution has been designed based on the results of incompany requirements and specification workshops conducted for a modular house builder as well as a manufacturer of dental products. Albeit a construction SME, the modular house builder recognised similar priorities compared to manufacturers and is thereby applicable to this study [22]. In the following, we present the design of both solutions in detail before applying the two-stage approach.…”

Section: Case Examplesmentioning

confidence: 99%

Configurable Solutions for Low-Cost Digital Manufacturing: a Building Block Approach

Kaiser

Ling

Yılmaz

et al. 2022

2022 IEEE 27th International Conference on Emerging Technologies and Factory Automation (ETFA)

View full text Add to dashboard Cite

The creation of digital manufacturing solutions at low cost is characterised by a successive development and a combination of disparate hardware and software elements. Modular building blocks address these limitations. This paper proposes and evaluates an explicit two-stage approach to develop configurable digital manufacturing solutions from bespoke solutions by using modular hardware and software building blocks. In the first stage, the solution is being made configurable through decomposition, while the second stage creates building blocks from resulting solution elements. For demonstration, the approach is applied to two tailor-made solutions. Resulting building blocks are then reused to create a new solution. We provide further insights by discussing challenges when applying our approach, and justifying its usage by qualitatively evaluating to what extent solution configurability can be achieved at low cost, which characterises the main objective of this study.

show abstract

“…However, most SMEs prefer simple and easy-touse solutions. This trend is primarily due to a lack of expertise and workforce on the SME teams and limited infrastructure [3]. More popular and additional uses of such monitoring solutions in these SMEs include knowing the operational states of their equipment to identify bottlenecks in production better, detecting idle machines, and planning staffing and resources more efficiently.…”

Section: Introductionmentioning

confidence: 99%

Ubiquitous Domain Adaptation at the Edge for Vibration-Based Machine Status Monitoring

Mukherjee

Parlikad

McFarlane

2023

IEEE Internet Things J.

View full text Add to dashboard Cite

This paper presents a Ubiquitous Domain Adaptation (UDA) and generalizability technique for vibrationbased automated machine status monitoring at the edge. The method significantly reduces the effects of signal noise artifacts and device/usage-specific vibration signatures using basic timefrequency domain signal operations and a lightweight ensemble of data-driven classifiers, allowing the method to be used for reliable domain-invariant status monitoring of motorized equipment. An experimental setup using vibration data from an air-cooled electric blender motor (source-domain) is used to train an automated machine state identification classifier that can identify the operating states of an eccentric rotating mass vibration motor (target-domain). Initial deployment of this method on target-domain motorized devices resulted in a machine status monitoring accuracy of at least 81.6% and a maximum training accuracy of almost 99% on known data of the source-domain and 91.49% for unseen data in the target-domain within an acceptable time frame. The performance of the proposed method is also comparable across platforms ranging from resource-constrained edge to a resource-rich cloud. This approach facilitates the use of noisy or uncalibrated sensor data in data-driven machine status monitoring tasks, therefore allowing for the development of reusable, low-cost monitoring systems that require meagre developmental effort, resulting in accelerated deployment times.

show abstract

Towards low-cost machine learning solutions for manufacturing SMEs

Cited by 9 publications

References 17 publications

Investigation of artificial intelligence in SMEs: a systematic review of the state of the art and the main implementation challenges

Investigation of artificial intelligence in SMEs: a systematic review of the state of the art and the main implementation challenges

Configurable Solutions for Low-Cost Digital Manufacturing: a Building Block Approach

Ubiquitous Domain Adaptation at the Edge for Vibration-Based Machine Status Monitoring

Contact Info

Product

Resources

About