Identification of evaluation criteria for algorithms used within the context of product development

Riesener, Michael; Doelle, Christian; Mendl-Heinisch, Michael; Klumpen, Niclas

doi:10.1016/j.procir.2020.02.207

Cited by 7 publications

(8 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This is not only a laborious course of action, the initial guess of suitable algorithms requires a certain amount of experience. (Lickert et al, 2021) To overcome this problem, there are several approaches to help finding the right ML algorithm for a given task (Waring et al, 2020;Riesener et al, 2020;Lickert et al, 2021;Gerschütz et al, 2021;Hashimi et al, 2015). A variety of the existing approaches can be summarized by the term AutoML.…”

Section: Related Workmentioning

confidence: 99%

“…If the ML algorithm does not match to the problem formulation at hand, it is highly unlikely that the ML algorithm is the optimal one to solve the problem. The literature distinguishes between regression, classification, and prediction (Riesener et al, 2020). However, the analysis of the research indicates, that the task can be further specified and assigned to the ML algorithms.…”

Section: Problemmentioning

confidence: 99%

“…Best known examples are Auto-Weka and Auto-Sklearn. (Waring et al, 2020) Riesener et al (2020 define evaluation criteria to facilitate the selection of ML algorithms in the context of product development. Therefore, the most used ML algorithms in this field have been evaluated by their accuracy, training time, computational effort, tolerance against erroneous and irrelevant parameters, online capability, and transparency.…”

Section: Related Workmentioning

confidence: 99%

“…The possibility to describe a problem to enable the comparison with a ML algorithm is not considered, thus complicating the application. For example, it is not stated how the established evaluation criteria by Riesener et al (2020) has to be applied to a given problem formulation. Moreover, the comparison between the criteria and the task and the ML algorithm respectively, requires expert knowledge and is only rational to a certain extent.…”

mentioning

confidence: 99%

See 3 more Smart Citations

A Pattern Language Approach to Identify Appropriate Machine Learning Algorithms in the Context of Product Development

et al. 2023

View full text Add to dashboard Cite

The product development process faces several challenges, such as an increasing and differentiated number of customer requirements, increasing product complexity, and shortened time-to-market. To address these challenges, the implementation of automation approaches in form of machine learning (ML) algorithms appears promising. However, companies lack the implementation of these approaches in their processes, inter alia due to inadequate knowledge and experience in this field. Therefore, the aim of this paper is to develop a structured formulized way of characterising ML algorithms, which can support non-experts in identifying the optimal algorithm to solve a given problem. First, existing approaches covering the determination of appropriate ML algorithms for a given task are examined. Based on this, a pattern language approach is introduced to characterise ML algorithms and problems, allowing matching to be performed to identify the most suitable one for a given task. Due to their broad application, the concept is demonstrated by creating patterns for decision trees and artificial neural networks. A study is conducted to prove that the proposed concept is appropriate to support the ML algorithm selection.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Problemmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

mentioning

confidence: 99%

See 2 more Smart Citations

A Pattern Language Approach to Identify Appropriate Machine Learning Algorithms in the Context of Product Development

et al. 2023

View full text Add to dashboard Cite

show abstract

“…Schönberg et al [17], seeing as distinct data arises from different stages of a product life cycle, defined product data management maintenance systems as adequate for being used in potentiating a better understanding of the conflicting factors related to a service or function. This type of innovation is a key success factor for a company and is of extreme importance for its R&D process [18][19]. Product development should also ensure an optimal performance/cost ratio while maintaining an effective risk management and intolerance to system flaws [20][21].…”

Section: Introductionmentioning

confidence: 99%

Case-based product development of a high-pressure die casting injection subset using design science research

Tojal¹,

Silva²,

Campilho³

et al. 2022

FME Transactions

View full text Add to dashboard Cite

High pressure die casting is widely used in metalworking industry. Many of the components and devices directly linked to the hot parts are subjected to severe wear. However, some design updates can enlarge the lifespan of these components and devices. This paper proposes to show how a sustainable maintenance can be obtained by focusing resources on the analysis of the critical problem and its consequent mitigation, whereas they are originated on the human, machine/process or supplier level. Design science research (DSR) was the iterative research methodology chosen to integrate this work. The completion of the objective with the implementation of this new injection subset concept brought extremely beneficial results, such as reduction of its consumption, reduction of acquisition cost and waste generated, and reduction of the intervention time during maintenance operations, as well as increase in equipment availability time.

show abstract

A conceptual MCDA-based framework for machine learning algorithm selection in the early phase of product development

Sonntag,

Pohl,

Luttmer

et al. 2024

Proc. Des. Soc.

View full text Add to dashboard Cite

Despite the potential to enhance efficiency and improve quality, AI methods are not widely adopted in the context of product development due to the need for specialized applications. The necessary identification of a suitable machine learning (ML) algorithm requires expert knowledge, often lacking in companies. Therefore, a concept based on a multi-criteria decision analysis is applied, enabling the identification of a suitable ML algorithm for tasks in the early phase of product development. The application and resulting advantages of the concept are presented through a practical example.

show abstract

Identification of evaluation criteria for algorithms used within the context of product development

Cited by 7 publications

References 15 publications

A Pattern Language Approach to Identify Appropriate Machine Learning Algorithms in the Context of Product Development

A Pattern Language Approach to Identify Appropriate Machine Learning Algorithms in the Context of Product Development

Case-based product development of a high-pressure die casting injection subset using design science research

A conceptual MCDA-based framework for machine learning algorithm selection in the early phase of product development

Contact Info

Product

Resources

About