Skip to main content

2025 | Buch

World Conference of AI-Powered Innovation and Inventive Design

24th IFIP WG 5.4 International TRIZ Future Conference, TFC 2024, Cluj-Napoca, Romania, November 6–8, 2024, Proceedings, Part I

insite
SUCHEN

Über dieses Buch

This book constitutes the proceedings of the 24th IFIP WG 5.4 International TRIZ Future Conference on AI-Powered Innovation and Inventive Design, TFC 2024, held in Cluj-Napoca, Romania, during November 6–8, 2024.

The 42 full papers presented were carefully reviewed and selected from 72 submissions. They were organized in the following topical sections:

Part I - AI-Driven TRIZ and Innovation

Part II - Sustainable and Industrial Design with TRIZ; Digital Transformation, Industry 4.0, and Predictive Analytics; Interdisciplinary and Cognitive Approaches in TRIZ; Customer Experience and Service Innovation with TRIZ.

Inhaltsverzeichnis

Frontmatter

AI-Driven TRIZ and Innovation

Frontmatter
LLM-Based Extraction of Contradictions from Patents
Abstract
Already since the 1950s TRIZ shows that patents and the technical contradictions they solve are an important source of inspiration for the development of innovative products. However, TRIZ is a heuristic based on a historic patent analysis and does not make use of the ever-increasing number of latest technological solutions in current patents. Because of the huge number of patents, their length, and, last but not least, their complexity there is a need for modern patent retrieval and patent analysis to go beyond keyword-oriented methods. Recent advances in patent retrieval and analysis mainly focus on dense vectors based on neural AI Transformer language models like Google BERT. They are, for example, used for dense retrieval, question answering or summarization and key concept extraction. A research focus within the methods for patent summarization and key concept extraction are generic inventive concepts respectively TRIZ concepts like problems, solutions, advantage of invention, parameters, and contradictions. Succeeding rule-based approaches, finetuned BERT-like language models for sentence-wise classification represent the state-of-the-art of inventive concept extraction. While they work comparatively well for basic concepts like problems or solutions, contradictions − as a more complex abstraction − remain a challenge for these models. Even PaTRIZ, the latest and complicated multi-stage approach to extract contradictions, delivers only mixed results. This paper goes one step further, as it presents a method to extract TRIZ contradictions from patent texts based on Prompt Engineering using a generative Large Language Model (LLM), namely OpenAI’s GPT-4. The existing annotated patent dataset “PaGAN” is used to demonstrate the LLM-capabilities for extracting TRIZ contradictions from the section “State-of-the-Art” of USPTO patents. Contradiction detection, sentence extraction, contradiction summarization, parameter extraction and assignment to the 39 abstract TRIZ engineering parameters are all performed in a single prompt using the LangChain framework. Our results show that “off-the-shelf” GPT-4 is a serious alternative to PaTRIZ. Comparing the text similarity of the GPT-4 extractions with the annotated sentences from PaGAN we reach a high F1-value of 0.93 using the BERTScore metric.
Stefan Trapp, Joachim Warschat
AI Based Search Engine to Deploy a TRIZ Pointer to Chemical Effects
Abstract
Pointers to effects are a group of TRIZ tools which helps the inventor to master a greater knowledge of scientific phenomena and laws, so to suggest him or her different directions to reach possibilities of solution. Pointers to geometric, chemical, and technological effects have been theorized, but only those to physical effects have ever had concrete developments at the research level and as commercial applications.
The aim of this work is twofold: on the one hand, to bring the pointer back to chemical effects (CE), recovering little-known texts that are difficult to find but also difficult to interpret, as they have never been translated from Russian. The other aim is to contextualize these tools in the light of the recent achievements of artificial intelligence technologies in the field of information retrieval. A combination of AI tools, as NER (Named Entity Recognition), RAG (Retrieval Augmented Generation) and LLM (Large Language Model) have been combined in order to identify chemical features from several chemical sources, to index documents in order to answer user’s questions, to interact with this Knowledge-Base by a chatbot and finally to generate a complete and standardized output.
A comparison is presented between recent commercial applications of AI and traditional pointers to CE from TRIZ literature. In this paper it is explained how the system works, which are the potentialities according to the AI technologies evolution and a comparative study between a SW infrastructure developed by the authors in collaboration with university spin-off software house and others current AI commercial players like GPT or Gemini based applications.
Davide Russo, Matteo Cattaneo, Simone Avogadri
Integrating Generative AI with TRIZ for Evolutionary Product Design
Abstract
Democratizing, scaling, and automating creativity and innovation are essential for boosting competitiveness, efficiency, and cost-effectiveness in product creation. This capability will be a key differentiator for organizations seeking a competitive edge and a catalyst for ensuring a better future for humanity. We demonstrate how the TRIZ methodology and Generative AI provide a solid foundation for solving product design challenges. We show how technology and human feedback enable the automated discovery of product design improvements. To prove this, we have developed a technological proof-of-concept solution that leverages the strengths of both TRIZ and Generative AI. This solution explores product reviews, identifies product strengths and weaknesses, and generates an initial palette of potential product design problems that can be iteratively refined to drive product improvement decisions. We provide the theoretical foundation by utilizing TRIZ’s systematic problem-solving approach and specific examples of generated data, creating a consistent basis for evaluating the tremendous potential of Generative AI and TRIZ in product design and improvement processes.
Marin Iuga, Stelian Brad
Harnessing Generative AI for Sustainable Innovation: A Comparative Study of Prompting Techniques and Integration with Nature-Inspired Principles
Abstract
Amidst growing environmental challenges and the imperative for sustainable solutions, this study explores how generative artificial intelligence (AI) can drive innovation in process engineering. It investigates the effectiveness of different prompting techniques and their integration with nature-inspired principles (NIP) in fostering sustainable innovation. The study employs a comparative methodology to assess the effectiveness of two distinct prompting techniques: basic and AI-automated prompting. It also examines the influence of integrating NIP derived from various natural ecosystems on the generated solutions. Experiments were conducted using a generative AI model and analysing the output, focusing on the number of unique and overlapping ideas. Furthermore, the quality of AI-generated solution concepts was evaluated using four parameters, such as feasibility, novelty, usefulness, and sustainability, each rated on a scale of 0 to 2. Three case studies within the process engineering domain were used, each representing a different problem-solving scenario. The results showed that the integration of NIP, particularly through the “one by one” strategy in AI-automated prompting, significantly increased the number of unique ideas compared to basic prompting, demonstrating its effectiveness in enhancing idea diversity and quality. Concepts generated through this approach also scored higher in novelty and sustainability, aligning with sustainable innovation goals. These findings have practical implications for developing innovative and sustainable engineering solutions, particularly in the early phases of design, offering insights into effective strategies for leveraging AI in eco-innovation.
Mas’udah, Pavel Livotov, Björn Kokoschko
Neuro-Symbolic AI-Driven Inventive Design of a Benzoic Acid Extraction Installation from Styrax Resin
Abstract
The extraction of benzoic acid from natural resins such as Styrax holds considerable industrial significance, given its widespread use in pharmaceuticals, food, and cosmetics. This study introduces an approach to enhance the extraction process by designing a novel installation in this respect. The design roadmap integrates Generative AI with neuro-symbolic AI algorithms. We employ a neuro-symbolic AI framework that merges AI’s generative capabilities for initial design conceptualization with symbolic reasoning, enriched with TRIZ principles and Complex Systems Design Thinking (CSDT) methodologies. This combination aids in navigating complex problem-solving scenarios and promoting an innovative solution. Environmental issues are integrated throughout the design process to ensure that the solution also meets eco-sustainability objectives. Results indicate that the novel design markedly enhances the extraction efficiency of benzoic acid, reduces energy consumption, and lowers waste production. The design’s adaptability for industrial applications has been validated, with future enhancements aimed at incorporating real-time monitoring AI systems for dynamic adjustments based on raw material variability.
Stelian Brad, Vasile-Dragoș Bartoș, Emilia Brad, Costan-Vlăduț Trifan
Enhancing TRIZ Contradiction Resolution with AI-Driven Contradiction Navigator (AICON)
Abstract
This paper presents the AI-driven Contradiction Navigator (AICON), a tool engineered to address the inherent limitations of the traditional TRIZ Contradiction Matrix. By integrating advanced AI technologies, AICON enhances the identification, mapping, and resolution of contradictions with advanced precision and context sensitivity. Central to its architecture is the incorporation of Retrieval-Augmented Generation (RAG) AI, enabling the system to access and utilize diverse knowledge domains dynamically. This capability allows AICON to identify new inventive principles and effectively cover previously unaddressed areas within the TRIZ matrix. The system’s architecture is designed to facilitate AI-driven data enrichment, adaptive contradiction mapping, and tailored solution generation, all within an iterative learning framework that continuously refines its problem-solving efficacy. Preliminary research outcomes highlight AICON’s success in discovering novel inventive principles and expanding the TRIZ matrix’s applicability with contextually relevant solutions. These advancements underline AICON’s potential to improve inventive problem-solving methodologies.
Stelian Brad, Emilia Brad, Alexandru Cîrlejan
Research on Disruptive Technology Prediction Methods Based on BERT Model and Graph Theory Analysis
Abstract
With the acceleration of the technological revolution, disruptive technologies have become a key factor in global technological competition. However, existing prediction methods are limited by single technology fields, semantic analysis limitations, and subjective factors, making it difficult to effectively predict these technologies. In this paper, we studied the current disruptive technology prediction methods using the ideal solution analysis and resource analysis tools of TRIZ theory and proposed a new prediction method. This method combines the BERT model and graph theory analysis for the first time, and it analyzes patent text, mines the inherent relationships between cross-domain technologies, extracts disruptive technology features, and evaluates them through expert judgments. This method fills the gap in existing research. Our method demonstrates unique innovation in cross-domain technology integration and can more accurately predict disruptive technologies. The research results show that the patent technologies selected after being fused perform excellently in terms of performance and advantages, verifying the scientificity and effectiveness of our research framework. This study provides a new and effective method for exploring and predicting disruptive technologies, which is expected to drive further development in related fields.
Zhongyou Wang, Jianhui Zhang, Rongjian Li, Xiangdong Guo
Research on the Identification and Analysis of Technological Opportunities Utilizing the BERT Model and MULTIMOORA Approach
Abstract
Product innovation is considered a crucial component of enhancing corporate competitiveness. Identifying technological opportunities plays a pivotal role in the success of a company’s research and development (R&D) activities. Technological opportunities are defined as the potential for technological advancements in specific areas. Historically, product innovation often relied on identifying singular technological opportunities. However, the current challenge lies in recognizing and leveraging a series of interacting multiple technological opportunities. To address this, our study introduces a novel approach based on the Transformer-based Bidirectional Encoder Representations from Transformers (BERT) model. This method transforms multiple technological objectives into generalized functional behaviors by restructuring them. An in-depth analysis of patent databases is conducted using the semantic search tool from Patsnap, extracting patent information related to these functional requirements. Subsequently, the data undergoes a transition from qualitative to quantitative analysis using spherical fuzzy sets. Finally, the quantified technologies are ranked for opportunities using the MULTIMOORA method, thus completing the identification of single or multiple interacting technological opportunities.
Jianguang Sun, Runze Miao, YuJuan Du, Delong Zhang
AI-Aided Resource Mining Method for Idealization-Driven Product Innovation
Abstract
The availability of resources in design plays a crucial role in innovative design. The importance of available resources increases as the solution to the problem approaches the Ideal Final Result (IFR). This paper introduces a method for mining resources in the product innovation process by integrating the Theory of Inventive Problem Solving (TRIZ) and AI, and demonstrates TRIZ’s application in various fields.
Firstly, through AI tools to obtain research hotspots to formulate product design goals, through the training of ChatGPT, with the help of AI tools to achieve the mining of internal resources. Additionally, an AI algorithm is employed to analyze patents and identify external available resources. We evaluate the value of these mined resources using a Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) comprehensive evaluation model that combines analytic hierarchy process with entropy weight method to select optimal available resources. Multi-criteria evaluation methods and AI technology are applied to resource mining and selection for product innovation.
Ranye Du, Jianguang Sun, Runze Miao, Delong Zhang
Use of AI in the TRIZ Innovation Process: A TESE-Based Forecast
Abstract
Currently, AI is mainly used in the TRIZ innovation process to find solutions to well-defined technical problems using TRIZ tools such as Inventive Principles, Standards and, occasionally, Functional Oriented Search (FOS). In practice, however, problem solving is usually the least time-consuming part of the innovation process, with most effort normally spent defining the overall goal of the innovation, identifying and analyzing the initial problem, selecting the best solution from the set of solutions found, and justifying its feasibility. Therefore, AI would be much more useful if it were introduced into these labor-intensive parts of the TRIZ innovation process as well, and undoubtedly AI developers will eventually try to automate the entire innovation process. The objectives of this paper are (1) to assess the current effectiveness of using AI in real TRIZ projects, (2) to predict the most likely sequence of future AI implementation in different parts of the TRIZ innovation process, and (3) to identify related challenges. The objectives are achieved by analyzing the composition and timing of various activities in a typical TRIZ project and applying the Trend of Decreasing Human Involvement to these activities, where a TRIZ project is considered a technological process that transforms an initial, poorly formulated problem into a viable solution/product. These results can be used by AI and TRIZ specialists to create a roadmap for integrating AI and TRIZ to produce a fully automated innovation process that is applicable to technical systems and, potentially, to business systems.
Oleg Abramov
Evaluating the Effectiveness of Generative AI in TRIZ: A Comparative Case Study
Abstract
The rapid advances in generative AI technologies have sparked a debate among researchers on their role in the innovation process, particularly regarding their problem-solving and idea-generation capabilities. While researchers theorise the potential of generative AI in conjunction with TRIZ (Theory of Inventive Problem Solving), evaluating its current state and understanding its practicality is equally critical. Hence, this paper provides evidence of generative AI’s ability to offer solutions in real innovation projects. Our exploratory study compares the results of an actual innovation project in a professional consulting-like setting using traditionally applied modern TRIZ tools against generative AI-assisted results for the same customer-defined problem. The comparison focuses on the solutions’ degree of similarity, depth, and breadth. Additionally, our research identifies the advantages, disadvantages, and feasibility of using generative AI in problem-solving and innovation projects. Our findings indicate that combining generative AI and TRIZ produces feasible, cross-domain preliminary conceptual directions with satisfactory scientific substantiation. Lastly, we recommend suitable use cases for innovation managers and TRIZ practitioners, highlighting how the TRIZ-GPT combination can save considerable time exploring preliminary concepts and idea generation during problem-solving.
Nikhil Phadnis, Marko Torkkeli
On Opportunities and Challenges of Large Language Models and GPT for Problem Solving and TRIZ Education
Abstract
The advent of GPT has caused a real revolution in many application contexts. Even the TRIZ community has had to face up to this new technology, questioning the possible integrations with traditional paths and tools. Many problem-solving experts have for some time been proposing specific prompts based on the methodology’s tools such as functional analysis, reconstruction of cause-effect relationships, identification of Resources, 40 inventive principles, etc., in order to support the problem solver, or even replace him altogether, during the inventive process. The free generation of LLM content has been applied for very different purposes such as, for example, to contextualize general purpose heuristics in specific domains, or as a search engine to answer technical questions, to suggest creative ideas or improve the formulation and redefinition of a problem, or finally to find connections between different application contexts.
This article proposes a critical analysis of the real effectiveness of these prompts according to the different needs of users.
The analysis was carried out using a software application that was developed in-house and for which a testing phase was conducted on a variegated sample covering both the academic and industrial fields, with more experienced users and users who have been approaching TRIZ for less time.
Simone Avogadri, Davide Russo
Challenges in Inventive Design Problem Solving with Generative AI: Interactive Problem Definition, Multi-directional Prompting, and Concept Development
Abstract
This paper explores the application of generative AI for systematic and inventive problem-solving in engineering design. Utilizing a multi-directional prompting approach, the study investigates the ability of AI chatbots to generate and evaluate innovative concepts based on numerous elementary solution principles. The research involved two sets of experiments with graduate and undergraduate students solving seven engineering design problems. The findings indicate that while generative AI can quickly generate a large number of ideas, it often overestimates the feasibility and usefulness of its solutions and tends to create overly complex designs. Comparisons of AI evaluations with those conducted by human participants revealed significant differences, highlighting the need for human oversight to ensure practical and contextually relevant outcomes. The experiments also revealed performance differences among various AI models, confirming a bias in self-assessment. Despite these challenges, integrating generative AI with multidirectional prompting using elementary inventive principles, primarily based on TRIZ methodology, proved effective in fostering innovative solutions. The study highlights the potential of simultaneously applying different AI models alongside human expertise to leverage the strengths of both large language models and human technical creativity.
Pavel Livotov, Mas’udah
The Evolving Landscape of TRIZ: A Generative AI-Powered Perspective
Abstract
The surge of Generative AI has revolutionized problem-solving, giving rise to innovative tools that unlock unprecedented solutions and cross-industry breakthroughs within the TRIZ methodology. This paper unveils five groundbreaking Generative AI- integrated tools designed to enhance innovation and problem-solving across diverse domains.
1.
Mechanism Oriented Search (MOS): Identifies and analyzes specific problem mechanisms, abstracting them for cross-industry comparison, facilitating the discovery of innovative solutions by applying insights from one field to challenges in another.
 
2.
Resource Innovator for Non-Engineering: Extends TRIZ to non-engineering fields, focusing on identifying and leveraging unique resources within domains like nursing, education, and communication, empowering users to uncover hidden potential.
 
3.
TRIZ FOS-Market Explorer: Facilitates the discovery and analysis of adjacent market opportunities by abstracting the primary function of a product or service and identifying similar functions across various industries, revealing potential new markets.
 
4.
Systematic Idea Generation: Employs detailed resource analysis and TRIZ principles to facilitate innovation within existing systems, categorizing resources and suggesting strategic modifications to components or processes.
 
5.
Function Redirector: Fosters innovation by redirecting functions and resources towards achieving goals in novel ways, deconstructing primary functions into auxiliary functions to stimulate creative problem-solving.
 
These tools collectively harness the power of Generative AI to revolutionize problem-solving and innovation across various sectors, offering structured analysis, imaginative recombination, and cross-disciplinary insights.
Tanasak Pheunghua
Backmatter
Metadaten
Titel
World Conference of AI-Powered Innovation and Inventive Design
herausgegeben von
Denis Cavallucci
Stelian Brad
Pavel Livotov
Copyright-Jahr
2025
Electronic ISBN
978-3-031-75919-2
Print ISBN
978-3-031-75918-5
DOI
https://doi.org/10.1007/978-3-031-75919-2