Aided Product Design Research Articles

Big Data and AI-Driven Product Design: A Survey

As living standards improve, modern products need to meet increasingly diversified and personalized user requirements. Traditional product design methods fall short due to their strong subjectivity, limited survey scope, lack of real-time data, and poor visual display. However, recent progress in big data and artificial intelligence (AI) are bringing a transformative big data and AI-driven product design methodology with a significant impact on many industries. Big data in the product lifecycle contains valuable information, such as customer preferences, market demands, product evaluation, and visual display: online product reviews reflect customer evaluations and requirements, while product images contain shape, color, and texture information that can inspire designers to quickly generate initial design schemes or even new product images. This survey provides a comprehensive review of big data and AI-driven product design, focusing on how big data of various modalities can be processed, analyzed, and exploited to aid product design using AI algorithms. It identifies the limitations of traditional product design methods and shows how textual, image, audio, and video data in product design cycles can be utilized to achieve much more intelligent product design. We finally discuss the major deficiencies of existing data-driven product design studies and outline promising future research directions and opportunities, aiming to draw increasing attention to modern AI-driven product design.

Classifying Books by Genre Based on Cover

A book cover can convey a lot about the content of the book. Despite the adage to not evaluate something based on outward appearances, we apply machine learning to see if we can, in fact, judge a book by its cover, or more specifically by its cover art and text. The classification was done considering three different aspects - cover image only, cover text only and both image and text in a multimodal approach. Image classification was done using transfer learning with Inception-v3. For text detection from the cover image, images were first converted to greyscale and different thresholds were applied to detect maximum text. This text was then vectorized and used to train a Multinomial Naïve Bayes model. We also trained custom CNNs for image and text modalities. For multimodal classification, we examine late fusion model, where the modalities are combined at decision level, and early fusion model, where the modalities are combined at the feature level. Our results show that the late fusion model performs best in our setting. We also observe that text is more informative with respect to genre prediction and that significant efforts need to be devoted to solve this image-based classification task to a satisfactory level. This research can be used to aid product design process by revealing underlying information. It could also be used in recommender systems and to help in promotion and sales processes for automatic genre suggestion.

Computer-aided product design of alternative solvents based on phase equilibrium synergism in mixtures

A systematic methodology is proposed that finds binary azeotropic mixtures as new alternative solvents for the extraction process of volatile aroma molecules widely used in perfume and cosmetic industries. We investigated the use of the reverse engineering approach with Computer Aided Product Design (CAPD) instead of the traditional ‘trial and error’ approach. First, the design problem is defined from the real functionalities of the classical solvents. They are translated into physicochemical properties and the bound values for each property are defined. The reverse engineering method coupled to CAPD consists in using optimization techniques for building molecular structures that match as best as possible the complete set of target physicochemical properties, thus defining for each candidate a performance index. Property values are evaluated by using group contribution methods of each molecular structure generated by CAPD tool or by using database values. Acknowledging the contradictory relationship between two selected physicochemical properties i.e, low boiling temperature and high flash point which is rarely found in pure components, binary azeotropic mixtures were studied in order to enhance the global performance of solvent candidates. Dimethyl carbonate (DMC) is an existing solvent for the extraction of odorous molecules from plants that exhibits a good ratio between the boiling temperature and the flash point. It is selected as the key component for designing binary azeotropic mixtures. DMC. The global performance of the binary azeotropic mixtures was verified by means of calculations of the vapour-liquid equilibrium and liquid – liquid equilibrium using Modified UNIFAC method as thermodynamic method.

Coupled Deterministic and Stochastic Modeling of an Industrial Multi‐Zone LDPE Autoclave Reactor

SummarySimulation aided product design of LDPE often requires deep understanding of polymer architectures produced in the industrial polymerization. Therefore, a hybrid‐stochastic approach is further developed that allows for the good description of the typical polymeric microstructure produced in a multi‐zone autoclave reactor. The here presented kinetic network goes beyond the basic polymerization scheme and accounts for the propagation of terminal double bonds as well as the mechanical degradation of polymer molecules due to shear stress. By including these reactions a good agreement between experimental and predicted molecular weight distribution is achieved. Furthermore, the comparison of individual contraction factors from experimental data and contraction factors calculated by a 3‐D random walker result in a good agreement.

Market Effects in Lifecycle Assessment: A Framework to Aid Product Design and Policy Analysis

Researchers have developed several methods to assess lifecycle environmental impacts of decisions in product design and policymaking. A major challenge is that whether impacts are reduced or exacerbated depends on market effects such as how the design change or policy influences the demand, use, and end-of-life of the relevant product and other products. However, little guidance is available to determine when market effects matter and how to model them. This paper identifies four categories of market effects and presents a framework to help researchers identify a priori whether these effects significantly influence environmental impacts and select an appropriate method.

Computer aided product design tool for sustainable product development

A computer aided product design (CAPD) tool is proposed that finds mixtures matching target properties. Genetic algorithm crossover and mutation operators are completed with insertion or deletion operators adapted for side branches. A new substitution operator is devised for cyclic molecules. The mixture fitness is evaluated by a weighted sum of property performances. Molecules are represented by molecular graphs. They are split into molecular fragments which are built from polyatomic groups. Molecules or molecular fragments can be fixed, constrained or left free for building a new molecule. Building blocks are chemical functional groups or bio-sourced synthons. A specific coding of hydrogen-suppressed atoms is devised that can be used with various property estimation models where atom connectivity information is required. Illustration is provided through three case studies to find levulinic, glycerol and bio-based derivatives as substitute for chlorinated paraffin, methyl p-coumarate ester solvent and blanket wash solvent, respectively.

Chemical enterprise model and decision-making framework for sustainable chemical product design

The chemical product substitution process is undertaken by chemical industries for complying with regulations, like REACH in Europe. Initially devoted to chemists, chemicals substitution is nowadays a complex process involving corporate, business and engineering stakeholders across the chemical enterprise for orienting the search toward a sustainable solution. We formalize a decision making process framework dedicated to the sustainable chemical product design activity in an industrial context. The framework aims at improving the sharing of information and knowledge and at enabling a collaborative work across the chemical enterprise stakeholders at the strategic, tactical and operational levels. It is supported by information and communication technologies (ICT) and integrates a computer aided molecular design tool. During the initial intelligence phase, a systemic analysis of the needs and usages enables to define the product requirements. In the design phase, they are compiled with the help of a facilitator to generate the input file of a computer aided product design tool. This multiobjective tool is designed to find mixtures with molecular fragments issued from renewable raw materials, and is able to handle environment-health and safety related properties along with process physicochemical properties. The final choice phase discusses the solution relevancy and provides feedback, before launching the product manufacturing. The framework is illustrated by the search of a bio-sourced water–solvent mixture formulation for lithographic blanket wash used in printing industry. The sustainability of the solution is assessed by using the sustainability shades method.

The importance and feasibility of “Reflectivity” as an index applicable for architectural acoustic design

A major part of room acoustics concerns sound control within a space. This traditionally involves reverberation control, music envelopment, optimization of speech comprehension and privacy, noise control, spatial enhancement, modal and reflection control for rooms used in recording, mixing, editing, mastering, and measuring sounds and hearing. In all of these examples, room acoustics is influenced by early reflections from the interior surfaces, objects and geometry. The impedance of a surface generally varies in relation to the incident angle of sound waves, therefore knowing the true reflectivity of surfaces would be complimentary to sound decay time and absorption coefficients. This presentation reviews current trends in the measurement of sound reflectivity as well as many of the challenges involved in developing an approved laboratory measurement methodology. Not only a potentially important tool for acousticians, architects and designers, but a formally adopted “Reflectivity Index” would aid product design, research, and education in room acoustics. Examples of comparative measurements of materials ranging from porous media to continuous or perforated surface assemblies are examined and how the range of results might be unified as a single metric.

Application of fuzzy logic-based environmental conscious QFD to rotary switch: a case study

General public and government agencies seriously view the environmental issues. Environmental issues have become an imperative concern for most companies in relation to modern product development. These activities stimulate the environmentally friendlier production technologies with a focus on minimization of environmental impact. Special procedures have to be formulated during the product development process to comply with recent green directives. In order to improve the focus on eco-designed products with regard to customer needs and cost considerations, this study presents the application of fuzzy eco-quality function deployment (QFD) to aid product design team in considering environmental concerns. This fuzzy multi-objective model considers the overall customer satisfaction and encourages enterprises to produce environmentally friendlier products. The case study describing the evolution of eco-friendly model of rotary switch has been presented. An optimal balance between environmental acceptability and overall customer satisfaction has been obtained as an outcome of the conduct of the study.

Modeling and optimization of product design and portfolio management interface

The paper presents modeling and analysis of product design and product portfolio management (PD–PM) domains interaction using an integrated simulation–optimization model. To represent the interactions, the product design phase is modeled as a discrete-scenario static system. The goal of this work is to develop a decision support framework that relies on product design–product portfolio management integration in order to aid product design planning and design execution. We utilize dependency matrix approach to illustrate domain relation between the product design and product portfolio management domains, and to facilitate their integration. Hence, the process integration model utilizes iterative effects, and their attendant processing duration and costs, to pattern domain interaction. An industrial case study is used to illustrate the application and utility of the proposed approach.

‘From the formal to the innovative’: The use of case studies and sustainable projects in developing a design process model for educating product/industrial designers

The BSc in computer aided product design (CAPD) course at the University of Wolverhampton was conceived as a collaborative venture in 1989 between the School of Engineering and the School of Art and Design. The award was at the forefront of forging interdisciplinary collaboration at undergraduate level in the field of product design. It has integrated engineering design from the former with industrial design from the latter, developing a mixture of traditional, technical and design skills using the computer. The current paper discusses the use of formal methods of delivery on the CAPD course together with the integration of industrial product design and engineering design from a case study and project approach. It progresses from a formal approach through a transitional stage to an innovative method of delivery and assessment. In providing the framework to this innovative approach the paper utilizes examples of students’ case studies, projects and design competitions that rely upon the use of three-dimensional (3D modelling) and both traditional prototyping and rapid prototyping techniques in developing the product design artefact. Combined with this is the utilization of a design process model that combines both product design and industrial design methodologies. The changing attitude of staff in the education of CAPD students, with respect to the integration of formal methods and innovative methods in case study and project scenarios, is also discussed.

Consumer preferences for forthcoming innovations: the case of high definition television

Examines consumer preferences for forthcoming technological innovations. Studies consumer response to different levels of technology over time. Looks at preferences for existing and intermediate technologies when future ones are expected. Focuses on technological consumer durables that are expected to evolve over time. The primary contributions of this research are: an understanding of how consumer preferences for a technology are affected by the time of adoption; an understanding of the value to consumers of obtaining different levels of a technology, at different points in time; and the utilization of parsimonious indices to assess consumer response to different levels of technology over time. An empirical examination is conducted for high definition television (HDTV). Using survey data, the study explores consumer preferences for HDTV, and for interim television technologies. Managerial implications to aid product design, and the timing of introduction of evolving technological innovations, are also discussed.

Computer aided product design: problem formulations, methodology and applications

A general methodology for Computer Aided Product Design (CAPD) with specified property constraints which is capable of solving a large range of problems is presented. The methodology employs the group contribution approach, generates acyclic, cyclic and aromatic compounds of various degrees of complexity (including isomers) and size (including polymers) and provides accurate and consistent estimation of properties. The application range of the methodology has been enlarged and the validity of the predictions improved through the incorporation of a new group contribution based method for estimation of properties. Also, with new problem formulations, it is possible to consider the formation of azeotropes and the miscibility of mixtures for the substitution of compounds, the design of polymers and the design of solvents for extraction operations involving vapor-liquid equilibria (VLE), liquid-liquid equilbria (LLE), solid-liquid equilibria (SLE) and gas solubility. Finally, a computer program based on the extended methodology has been developed and the results from five case studies highlighting various features of the methodology are presented.