The current ways of coping with uncertainty such as changes during product design or use have been through methods such as easy restructuring (e.g., modularity with buffer in interface definition), by overdesign and so on. The present investments on maintaining products in the economy for “as long as possible” is challenging these strategies from a cost and environmental perspective. Moreover, these strategies often lead to highly overdesigned products. An alternative strategy is to introduce features in a design, called “resilient objects”, which are able to absorb such uncertainties without wasteful overdesign of other parts. By applying a ‘text-mining’ approach on patents, this paper has identified 5,552 candidates for such resilient objects that can be recombined and inserted in regions of the product that are likely to be most affected by current and future uncertainties. The application of resilient objects is demonstrated on a case study (a cooling system for battery electric vehicles). The case study highlights the ability of these objects to 1) significantly increase protection against uncertainties without the need for restructuring, 2 ) reduce the risk for overdesign and 3) dampen effects of change propagation.

Patents are an invaluable source of data that can be beneficial for Engineering Design (ED). Patenting is one of the main means for disclosing the inventive process. For this reason, the description of the problem solved should also be included in any patents.

The ED literature lacks a proper definition of a problem, resulting in a fragmented scenario. Prior studies have employed Text Mining (TM) to extract problems from patents. We argue that TM can assist ED researchers in understanding how problems are articulated in text. Based on the literature, we propose two hypotheses: (1) problem-related text exhibits a negative sentiment polarity compared to other sections of patents; (2) problem-related keywords identified in the literature are predominantly used to describe problems rather than other aspects.

We analyse Japanese patents to validate our hypotheses, since they explicit Problem and Solution in the abstract. Finally, we compare our results with a set of problem-related sentences extracted from USPTO patents.

Our study reveals a higher positive sentiment in problem-related sentences compared to solution-related ones and highlights the inadequacy of using problem-related keywords alone to differentiate between the two.

Digital media are a means to deliver products and services, but also a channel to interact with consumers and a source of information on users’ preferences. Data shared by customers on the web, the User-Generated Content (UGC), can give entrepreneurs a detailed perspective of the market. This work examines an application of Natural Language Processing techniques on UGC to discover insights on users' opinions. We collected more than 13.000 reviews of software from digital stores and review website to gather information on the customers’ perspective and their response to a given marketing strategy in two case studies on digital product's launch. The objective is to give support to two Italian companies in the process of business model development through data-driven evidence. We aim to discover who are the users and which are their needs using a lexicon-based approach to mine unstructured text. The results provide qualitative and quantitative descriptions of the market segments. We propose a method to examine UGC and to explore customers’ behavior on social media. The findings helped managers for the development of their business model, enhancing an informed decision-making process.

COVID-19 pandemic has continued to pose a challenge to the society for almost three years, adversely affecting all segments of population in a scale unseen in the recent decades. Over the course of COVID-19 pandemic, many people have lost their jobs and income. These social and economic impacts have disrupted the market, potentially altering people's attitudes towards different product features. Therefore, this paper investigates the changes in customer preferences on various features of different products, before and after COVID-19 pandemic, using online review analysis. The proposed framework consists of four stages. Firstly, product review data is collected and preprocessed. Secondly, customer interest in product features is explored using latent Dirichlet allocation. Thirdly, customer sentiment for these features is analyzed with Valence Aware Dictionary and sEntiment Reasoner. Finally, the importance of each feature is calculated based on interpretable machine learning. The proposed method is tested on two real-world datasets – smartphone and laptop reviews. The result reveals the changes in customer sentiments and preferences for product features, thus helping companies quickly establish strategies in rapidly changing market environments.

It is important for organizations to balance exploration and exploitation in order to respond quickly and sustainably to the needs of society and users in a rapidly changing business environment. However, there is insufficient research on design methods to balance exploration and exploitation in product design, and a method to objectively identify the product groups to be balanced has not yet been established. In this paper, on the basis of the characteristics of exploration and exploitation in design, a cluster analysis using functional and attribute distances between products is proposed. To validate the proposed method, it was applied to past product cases in which the relationship between exploration and exploitation was known. The results showed that in the cases of cameras, in addition to known product groups forming large clusters, reasonable minor classifications that had not been identified were also obtained. This indicates that the proposed method is capable of analyzing reasonable clusters in the cases and is potentially effective in identifying product groups taking into consideration the relationship between exploration and exploitation.

The research presented in this paper explores a novel method for assessing the effects of biophilic restorative experiences on designers’ cognition by combining the use of physiological, neurocognitive and semantic measures. A total of 12 engineering graduate students participated in a three-step pilot experiment that consisted of (1) a stressor (the Trier Social Stress Test), (2) a destressing intervention (biophilic sound experience), and (3) a design task. Heart rate variability (HRV) was used to track subjects’ autonomic nervous system (ANS) activity. Functional near-infrared spectroscopy (fNIRS) was used to track patterns of brain activation in subjects’ prefrontal cortex (PFC). Changes in design quality were assessed by the semantic space they explored, measured through a natural language processing (NLP) technique. Preliminary findings suggest that an auditory biophilic restorative experience can change designers’ bodies, brains, and minds. Results from this pilot study encourage further exploration of the use of exposure to nature-based stimuli as a method to help enhance engineering design cognition.

The requirements space is increasing due to non-functional areas such as security, resilience and sustainability gaining in importance. This creates a complex and dynamic space which makes it hard for engineers to take good data driven design decisions. Increasing the quality of design decisions allows to better set up development projects and develop more successful products and services. The design can most heavily be influenced in the early design phases, where design flexibility is high and resource commitment is low. Unfortunately, the system knowledge is also low in early phases. The Engineering Graph is a concept that connects data from different internal and external sources. It allows to connect product data stored in Product Lifecycle Management systems with system models and also add external sources from the Wikimedia Knowledge Graph, World Health Organization and World Bank. This interconnected data allows the support of engineers in managing the complex and dynamic requirement space and provide high system knowledge in the early design phases to support design decisions.

Currently, engineers need to manually analyse requirement specifications for determining parameters to create geometries in generative engineering. This analysis is time-consuming, error-prone and causes high costs. Generative engineering tools (e.g. Synera) cannot interpret natural language requirements directly. The requirements need to be formalised in a machine-readable format. AI algorithms have the potential to automatically transform natural language requirements into such a formal, machine-readable representation. In this work, a method for formalising requirements for generative engineering is developed and implemented as a prototype in Python. The method is validated in a case example using three products of an automotive engineering service provider. Requirements to be formalised are identified in the specifications of these three products, which are used as a test set to evaluate the performance of the method. The results show that requirements for generative engineering are formalised with high performance (F1 of 86.55 %). By applying the method, efforts and therefore costs for manually analysing requirements regarding parameters for generative engineering are reduced.

In product development, it is of great importance that a complete, unambiguous, and, as far as possible, contradiction-free target system is defined. Requirements documents of complex systems can contain several thousand individual requirements, derived in an interdisciplinary manner and written in natural language by many different stakeholders. Hence, errors, in the form of contradictions, cannot be completely avoided in these documents and today they must be corrected manually with high effort.

This paper presents an important building block for automated contradiction detection and quality analysis of requirements documents. We discuss the necessary identification of conditions in requirements and the extraction of the verbal expressions associated with condition and effect, respectively. We applied and analyzed natural language processing methods based on grammatical versus machine learning models. The models have been applied to 1,861 real-world requirements. Both approaches generate promising results, with an accuracy partly over 98%. However, in structured specification texts, a grammatical model is preferable due to lower effort in preprocessing and better usability.

With the growth of online marketplaces and social media, product designers have been seeing an exponential growth of data available, which can serve as an extremely valuable source of information communicated from customers without geographical limitations. The data will reveal customers’ preferences, which can be expensive and slow to obtain via traditional methods such as survey and questionnaires. While existing methods in the literature have been proposed to extract product information and make inference from online data, they have limitations, especially in providing reliable results and in dealing with data sparsity. Therefore, this paper proposes a method to conduct an Important-performance analysis from online reviews. The major steps of this method involve using latent Dirichlet allocation (LDA) to identify product attributes, using IBM Watson Natural Language Understanding tool to perform aspect-based sentiment analysis, and using XGBoost model to infer product attribute importance from the collected dataset. In our case study, we have collected over 150,000 text reviews of more than 3,000 laptops from Amazon.

Standards are an important source of knowledge in product development. Due to the increasing digitization of the product development process, standard development organizations aim to develop machine-actionable standards that automatically enforce operations in output devices. However, the current representation format in PDF or XML does not meet the requirements of machine-actionable standards. This paper examines existing approaches towards the representation of XML data in knowledge graphs and their transferability towards the domain of digital standards. Based on these approaches, the paper aims to develop and validate a concept for transferring standard content from XML format to a graph-based representation, using the example of formulas. For this purpose, a concept for the automatic identification, extraction and modeling of formulas will be presented. Afterwards, the concept is validated using the example of DIN ISO 281 whereas a chatbot application serves as conversational user interface. It is proven, that knowledge graphs are suitable for the representation of machine-actionable standard content. Future work will investigate the abstraction towards a general approach as well as further information objects of standards.

The representation of the product use context is a well established design practice in Engineering Design. Recently, design theory is studying the product interaction involving several cognitive aspects such as the possible conditions in which a wrong interaction occurs. The aim of this paper is to find a quantitative evidence of the causes of these misuses. In particular, this study focuses on the detection of bad design and biases.

In this paper, we propose a method that helps to the automatic detection of bad design and biases from patents. The method is based on an approach that defines syntactic rules to detect sentences containing these artifacts. These rules are defined based on an exploratory analysis of the explicit mention of “bad design” and “bias” and then, tested with multiple experiments on a sample of patents. The results give a first quantitative evidence of the presence of bad design and biases in patents and consequently of their importance in the design theory. In particular, it is provided a fine grain analysis of the linguistic structure of sentences containing these artifacts helping designers in detecting automatically them from patents.

Data, information and knowledge are strongly involved in Engineering Design (ED) process. Despite the crucial role played by data in the design process, there is a lack of studies about how different data are used and generated by the various phases of the ED process. This study is a first attempt to fill this gap by mapping which data types are involved in the different ED phases from a research perspective.

In order to achieve this objective, we used a methodology based on Text Mining. Firstly, we retrieve a corpus of scientific papers related to ED; then, we build two lexicons to recognize ED phases and data types; finally, we collect these entities within ED papers and map the relations between them.

The methodology application allows the building of a network graph for visualizing the relations among data lexicon and ED lexicon. Then, we investigate the specific relations among data types and ED phases by building a heatmap to investigate data types from 3 different perspective.

The insight coming from our analysis shows that ED studies have a great potential in the usage of many data sources, but also that there exist some gaps to be solved in order to reach a more effective data usage in the context of ED.

Engineers often need to discover and learn designs from unfamiliar domains for inspiration or other particular uses. However, the complexity of the technical design descriptions and the unfamiliarity to the domain make it hard for engineers to comprehend the function, behavior, and structure of a design. To help engineers quickly understand a complex technical design description new to them, one approach is to represent it as a network graph of the design-related entities and their relations as an abstract summary of the design. While graph or network visualizations are widely adopted in the engineering design literature, the challenge remains in retrieving the design entities and deriving their relations. In this paper, we propose a network mapping method that is powered by Technology Semantic Network (TechNet). Through a case study, we showcase how TechNet’s unique characteristic of being trained on a large technology-related data source advantages itself over common-sense knowledge bases, such as WordNet and ConceptNet, for design knowledge representation.

Patent analysis is a popular topic of research. However, designers do not engage with patents in the early design stage, as patents are time-consuming to read and understand due to their intricate structure and the legal terminologies used. Manually produced graphical representations of patent working principles for improving designers’ awareness of prior art have been demonstrated in previous research. In this paper, an automated approach is presented, utilising Natural Language Processing (NLP) techniques to identify the invention working principle from the patent independent claims and produce a visualisation. The outcomes of this automated approach are compared with previous manually produced examples. The results indicate over 40% match between the automatic and manual approach, which is a good basis for further development. The comparison suggests that the automated approach works well for features and relationships that are expressed explicitly and consistently but begin to lose accuracy when applied to complex sentences. The comparison also suggests that the accuracy of the proposed automated approach can be improved by using a trained part-of-speech (POS) tagger, improved parsing grammar and an ontology.

There have been growing uses of semantic networks in the past decade, such as leveraging large-scale pre-trained graph knowledge databases for various natural language processing (NLP) tasks in engineering design research. Therefore, the paper provides a survey of the research that has employed semantic networks in the engineering design research community. The survey reveals that engineering design researchers have primarily relied on WordNet, ConceptNet, and other common-sense semantic network databases trained on non-engineering data sources to develop methods or tools for engineering design. Meanwhile, there are emerging efforts to mine large scale technical publication and patent databases to construct engineering-contextualized semantic network databases, e.g., B-Link and TechNet, to support NLP in engineering design. On this basis, we recommend future research directions for the construction and applications of engineering-related semantic networks in engineering design research and practice.

The outbreak of the coronavirus disease not only caused many deaths worldwide but also severely affected the development of the global economy, such as supply chain disruptions, plummeted demand, unemployment, etc. These social changes have led to changes in customers' purchasing patterns. Therefore, it is more important than ever for manufacturers to quickly identify and respond to changing customer purchasing patterns and requirements. However, few studies have been done on dynamic changes in customer preferences for product features following COVID-19 spread. This study aims to investigate the dynamic change of customer sentiment on product features following COVID-19 through sentiment analysis based on online reviews. The proposed methodology consists of two main processes: feature extraction and sentiment analysis. After finding a specific feature of the product through feature extraction, the words used to mention the feature in the review were analyzed for sentiment analysis of customers. To demonstrate the methodology, a case study is conducted using new and refurbished smartphone reviews to investigate the dynamic changes in customer sentiment during COVID-19.

This study attempted to explore how pragmatic and hedonic values are influenced by the level of technology and what particular functions have to be considered in the context of smart technology- driven design in terms of Pragmatic Value (PV) and Hedonic Value (HV). An on-line questionnaire survey was developed to answer the research questions. A total of 104 respondents participated in the survey. As target product for the study, analog watch and smart watch were selected as representative of low and high technology respectively. Semantic Differentials on PV and HV were used and expected functions were investigated via an open question. The results indicate that there are some differences between analog and smart watches in terms of PV and HV. Regarding expected functions, significant differences were identified in the study. The findings from the study could provide a better understanding of the relationship between PV and HV in terms of level of technology. If it is considered in product development process, it may contribute to an increase of user satisfaction with smart- technology based product and service.

As design and design thinking become increasingly important competencies for a modern workforce, the burden of assessing these fuzzy skills creates a scalability bottleneck. Toward addressing this need, this paper presents an exploratory study into a scalable computational approach for design thinking assessment. In this study, student responses to a variety of contextualized design questions – gathered both before and after participation in a design thinking training course – are analyzed. Specifically, a variety of text features are engineered, tested, and interpreted within a design thinking framework in order to identify specific markers of design thinking skill acquisition. Key findings of this work include identification of text features that may enable scalable measurement of (1) user-centric language and (2) design thinking concept acquisition. These results contribute toward the creation of computational tools to ease the burden of providing feedback about design thinking skills to a wide audience.

The importance of affordance in Engineering design is well established. Artifacts that are able to activate spontaneous and immediate users’ reactions are considered the outcome of good design practice.

A huge effort has been made by researchers for understanding affordances: yet these efforts have been somewhat elusive. In particular, they have been limited to case studies and experimental studies, usually involving a small subset of affordances. No systematic effort has been carried out to list all known affordance effects. This paper offers preliminary steps for such an ambitious effort.

We propose a set of three different approaches of Natural Language Processing techniques to be used to extract meaningful affordance information from the full text of patents: 1) a simple word search, 2) a lexicon of affordances and 3) a rule-based system.

The results give in-depth measures of how rare affordances in patents are, and a fine grain analysis of the linguistical construction of affordances. Finally, we show an interesting output of our method, that has detected affordances for disabled people, showing the ability of our system to automatically collect design-relevant knowledge.