Ontologies and AI techniques for data integration and decision making in product design

The table of contents is as follows.

4.1 Introduction
  4.1.1 Product Design and Obsolescence Problem
  4.1.2 Current Status of Product Design and Obsolescence Management
  4.1.3 Ontology and Its Utilization
4.2 Framework and Work Process
4.3 Ontology Development, Optimization, and Utilization for Data Integration
  4.3.1 Ontology-Based Data Integration
  4.3.2 Ontology Development
  4.3.3 Ontology Optimization
     4.3.3.1 Structures of Ontologies
     4.3.3.2 Relations Between Ontologies
     4.3.3.3 Ontology Clustering
  4.3.4 Ontology Utilization
     4.3.4.1 Ontology-Based Data Integration
     4.3.4.2 Ontology-Based Product Design
4.4 Case Studies
  4.4.1 Ontology-Based Decision-Making Support in Product Design
     4.4.1.1 Ontology-Based Product Parameter Adjustment
     4.4.1.2 Ontology-Based Material and Machine Selection
  4.4.2 Ontology-Based Knowledge Representation and Decision Support for Managing Product Obsolescence
     4.4.2.1 Ontology Representation for Obsolescence Knowledge
     4.4.2.2 Knowledge Base for Obsolescence Management
     4.4.2.3 Framework of Obsolescence Management Information System
     4.4.2.4 Obsolescence Management Cost Analysis
4.5 Conclusions and Future Work

About the Product Planning Process

The product planning process, which is an important step in the development of a new product or improvement of an existing product, consists of the following steps

1. Market research and understanding customer needs: The first step in product planning is market research and understanding customer needs. Here, market trends and competitive information are gathered to understand customer requirements and needs. This will help you understand the direction of the product and the market requirements, and lay the foundation for your planning.
   Product concept development: Based on market research and customer requirements, product concepts are developed. This includes product features, functions, design concepts, etc. Multiple concepts are generated, evaluated and compared.
2. Requirements Definition and Design: After the concept is finalized, product requirements are defined. Requirements definition includes product functionality, performance, quality standards, regulatory requirements, cost constraints, etc. Based on the requirements definition, the product is designed to embody the product concept and structure.
3. Technical feasibility assessment: Evaluates the technical solution to meet the product requirements. This includes evaluating technical feasibility and manufacturing processes, identifying resources and capabilities, and evaluating technical risks and constraints to inform decision making.
4. Prototype Development and Evaluation: After technical feasibility is confirmed, prototypes are developed. The prototype is a tool for fleshing out the product’s functionality and design, and for evaluation and testing. Based on the evaluation results of the prototype, the design is modified and improved.
5. Detailed product design and manufacturing plan: The detailed design of the product is performed. This includes product structure, materials, manufacturing processes, quality control plan, etc. Detailed plans for the manufacturing process are developed and preparations for production are made.
6. Product Release and Market Launch: The final product is manufactured and introduced to the market. Product quality control and quality assessment are performed to prepare the product for release to the market. After market launch, monitor and provide feedback on the product and make any necessary improvements or adjustments.
7. Marketing Strategy and Sales Plan: Develops marketing strategy and sales plan for the product. Prepare the product for market launch by developing marketing plans, advertising strategies, and selecting sales channels.
8. Prototyping and evaluation: Prototype the designed product and create the actual product. Evaluate the prototypes to verify product performance and quality, and modify or improve the design based on the evaluation results of the prototypes.
9. Production Planning and Supply Chain Development: Develop a production plan for the mass production of the product and establish a supply chain. Plan production processes, procurement of materials, factory locations, and manufacturing schedules to ensure efficient production.
10. Marketing Strategy and Sales Plan: Develop marketing strategy and sales plan for the product. Prepare for product launch by developing marketing plans, advertising strategies, and selecting sales channels.
11. Monitoring and improvement: Monitor market response and customer feedback after the product is launched. Identify product improvements and issues, make necessary modifications and improvements, and track market changes and competitive trends to ensure continuous product improvement and competitiveness.

The product planning process in the manufacturing industry is an important process to understand market requirements and customer needs, and to develop and improve products accordingly. Data analysis and AI technologies can be used at each stage of this process to help understand market trends, analyze customer needs, optimize product design, and improve the efficiency of the production process.

B to C (Business to Consumer), in which companies provide products and services to individual consumers, and B to B (Business to Business), in which transactions and services are provided between companies. In B to C, (1) understanding demand and market research, (2) clear branding and differentiation, (3) marketing and customer experience, (4) quality and reliability, and (5) building customer relationships and customer loyalty are important, while in B to B, (1) trust and relationship building, (2) value proposition and competitive edge, (3) marketing and In B to B, (1) trust and relationship building, (2) value proposition and competitiveness, (3) marketing and segmentation, (4) quality and support, and (5) long term perspective and partnership become important, and the above processes need to be developed from a different perspective.

In this article, we will focus on decision making in the product design process. In the following, we first redefine the decision-making process in the product design process.

About the Decision-Making Process in the Product Design Process

Decision-making in the product design process is to select the optimal design while considering factors such as product quality, functionality, and cost-effectiveness, and is carried out in the following steps.

1. Goal setting: Define the goals of the product design. This includes product functionality, performance, quality criteria, cost constraints, etc. By setting specific goals, the metrics on which decisions are based can be clarified.
2. Generate design alternatives: Generate multiple design alternatives. This includes creating design alternatives that consider factors such as different materials, structures, dimensions, and manufacturing processes, and that leverage the knowledge and experience of the design team and experts to generate a wide range of options.
3. Establish evaluation criteria: establish criteria for evaluating design alternatives. This includes the importance and prioritization of target factors (e.g., functionality, quality, cost, etc.). Establishing evaluation criteria will facilitate comparison and selection of designs.
4. Evaluation and Analysis: Evaluate and analyze design alternatives. This includes numerical simulation, prototyping, and analysis of experimental results. Based on the evaluation results, the performance, cost, and risk of each alternative are compared to identify the superior design.
5. Decision making: Select the optimal design, taking into account the evaluation results. This includes numerical comparisons and decisions based on the evaluation results, as well as discussions with the engineering team and stakeholders. Multiple perspectives and opinions are comprehensively considered to make the final decision.
6. Monitoring and Feedback: After a decision has been implemented, the product performance and market changes are monitored and feedback is incorporated. This facilitates improvement of the product by modifying the design and decision-making process as necessary.

For these processes, it is possible to generate design alternatives, set evaluation criteria, and support the process of evaluation and analysis, and it is also possible to support more rational decision making through data-driven decision making and the use of predictive models to complement risk identification and future performance prediction.

Specific examples of ontology and AI technology applications are described below.

Ontology-based Decision Support and AI Technologies in Product Design

Ontology-based decision support and the use of AI technology in product design are important methods to support more efficient and accurate decision making. They are expected to be used in areas such as knowledge management, parameter design and optimization, manufacturing process optimization, design generation and evaluation, and risk assessment and decision support. Specific methods are described below.

• Knowledge Management: Using ontology-based knowledge representation to incorporate information such as product requirements, functions, constraints, materials, and components into an ontology to systematize and organize knowledge related to product design, thereby centralizing critical knowledge in product design and improving reusability and shareability.
• Problem Definition and Analysis: Ontologies can be used to define and analyze product design problems, and because knowledge is structured by ontologies, the essence of the problem can be clearly understood and appropriate solutions can be found.
• Parameter Design and Optimization: AI technology will be used to support product parameter design and optimization, combining ontology-based knowledge representation with AI models to enable optimal parameter design based on product characteristics and requirements AI models can learn from large amounts of data and learn from simulation results and are responsible for predicting and optimizing product performance and quality.
• Manufacturing Process Optimization: AI technology will be used to help optimize the manufacturing process of a product. This is accomplished by incorporating information and constraints related to the manufacturing process into an ontology and using AI models to design and control the optimal manufacturing process. This will improve product quality and maximize production efficiency.
• Design Generation and Evaluation: AI technology will be used to support product design generation and evaluation. Combining ontology-based knowledge representation with AI models, the AI models learn knowledge about the aesthetics and functional requirements of a design, suggest good design candidates, and automatically generate design alternatives, while using the ontology’s knowledge about the product’s functions and constraints to evaluate those designs. The design is then evaluated by using knowledge of the ontology’s product functions and constraints.
• Risk Assessment and Decision Support: AI technology is used to support risk assessment and decision making in product design. Using ontology-based knowledge representations of product requirements, constraints, and design parameters, combined with AI models, predict potential risks and defects and propose appropriate countermeasures and fixes. This will enhance the risk management and decision-making process.

As described above, ontology-based decision support and the use of AI technology are effective methods in product design. It is expected to be used in areas such as knowledge management, parameter design and optimization, manufacturing process optimization, design generation and evaluation, and risk assessment and decision support.

Ontology-based product parameter tuning and application of AI technology

The application of AI technology to ontology-based product parameter adjustment will enable decision support to optimize product performance and quality. The following describes the specific methods used.

• Parameter optimization: AI technology can be used to optimize product parameters. This could be, for example, using AI models to find optimal parameters for product design parameters or manufacturing process conditions, or combining ontology-based knowledge with AI models to explore optimal parameters for product characteristics and requirements to support decision making.
• Performance Prediction and Simulation: AI technologies can be used to predict and simulate product performance. Considering product characteristics, components, and environmental conditions, AI models can be used to predict performance, thereby evaluating product behavior for different parameter settings and changes and supporting decision making to select optimal parameters.
• Multi-objective optimization: In some cases, multiple objectives and constraints are considered when adjusting product parameters, and AI techniques are applied to optimize multiple objective functions and constraints, combining ontology-based knowledge with AI models to balance different objectives and search for optimal parameter settings. AI models can support decision making by searching for optimal parameter settings while balancing different objectives.
• Real-time data linkage and feedback loops: Leveraging AI techniques, product performance data can be collected in real-time and combined with ontologies for parameter tuning and decision making. The collected data can be fed back into the AI model to evaluate changes in product performance and the effects of adjustments for further optimization.

The use of AI technology enhances decision support in ontology-based product parameter tuning by letting AI handle parameter optimization, performance prediction and simulation, multi-objective optimization, real-time data linkage and feedback loops, and more Enables efficient and accurate parameter tuning and decision making.

Ontology-based material and equipment selection with AI

Ontology-based material and equipment selection can also be combined with AI technology to provide more efficient and accurate decision support. Specific methods are described below.

• Data-driven material and equipment selection: AI technology is used to learn the characteristics and performance of materials and equipment from large amounts of data. This can be done, for example, by learning from historical product data, customer feedback, and market trends to identify superior materials and equipment, thereby enabling data-driven decision making in material and equipment selection.
• Prediction and optimization of properties: AI technology can be used to predict and optimize the properties of materials and equipment. characteristics can be accurately evaluated to find the best options.
• Consideration of constraints: Constraints (e.g., budget, production requirements, environmental regulations, etc.) exist in the selection of materials and equipment. This would involve using AI models to identify options that satisfy constraints, for example, searching for the best option under budget constraints.
• Real-time data linkage: AI technology will be leveraged to provide real-time data linkage. Data on the performance and degradation status of products and materials will be collected in real time and fed into AI models to enable decision making based on the most up-to-date information. Combining ontological knowledge with real-time data will enable more accurate selection and decision making.

The use of AI technology will enhance decision support in ontology-based material and equipment selection. Faster and more accurate selection and decision-making will be possible by letting AI handle data-driven selection, property prediction and optimization, consideration of constraints, and real-time data linkage.

Risk Assessment and Decision Making in Product Design Using AI Technology

Risk assessment and decision making in product design using AI technology is important to improve product safety, reliability, and quality. The following describes how to conduct risk assessment and decision making using AI technology.

• Data Analysis and Predictive Model Building: AI models are built by learning from historical product data and related data. For example, we leverage product failure data, quality data, customer feedback, etc. AI models identify trends and patterns in the data and build predictive models to forecast product risk.
• Risk assessment and ranking: Using the predictive models, AI models assess the product’s risk, such as the likelihood of abnormal product behavior or failure, and the risk of quality issues, and rank the degree of risk. This allows for identification and prioritization of high-risk items and elements.
• Propose risk mitigation measures: AI models can identify risk factors and propose mitigation measures such as product design or process improvements. AI models can leverage historical data and knowledge bases to predict the effectiveness of risk mitigation measures and make optimal recommendations.
• Simulation and Validation: AI technology can be used to simulate and validate risk mitigation measures. Simulate how product design and process changes affect risk mitigation and evaluate the results. We also validate the results against actual products and processes. This helps to confirm the effectiveness of risk mitigation measures and inform decision making.
• Real-time monitoring and feedback: Real-time monitoring of product operations and usage data is fed back into the AI model. This allows us to detect risk factors and abnormal behavior of products and take countermeasures as early as possible. Feedback data is also used to improve prediction models to support more accurate risk assessment and decision-making.

Risk assessment and decision making using AI technology contributes to enhanced risk management and quality improvement in product design. It combines data analysis and predictive model building, risk assessment and ranking, risk mitigation recommendations, simulation and validation, and real-time monitoring and feedback to achieve more effective decision making.

Ontology-based knowledge representation and decision support for product aging management

The combination of ontology-based knowledge representation and AI technology can also further enhance decision support for product aging management. This can be done by utilizing AI technologies to analyze, predict, and automate large amounts of data. The following are examples of ontology-based knowledge representation and the use of AI technologies.

• Data collection and analysis: Data on product status and performance, such as sensor data and monitoring data, are collected and analyzed using AI technology; AI models learn patterns and trends in the data, build aging indicators and prediction models, and combine them with ontologies to more accurately represent the meaning and relationships of the data accurately represent and support decision making for aging management.
• Maintenance scheduling: AI technology can be used to optimize maintenance schedules based on product aging status and predictive models. By combining ontological knowledge with AI models, optimal maintenance planning can be realized, taking into account product characteristics, requirements, and maintenance resource constraints.
• Risk assessment and risk management: AI technology can be used to help assess and manage risks due to aging. Combining information on ontology and AI models, potential risks due to product aging can be identified and countermeasures and risk management strategies can be developed to address them. AI technology can learn from historical data and domain knowledge to improve the accuracy of risk assessment and risk prediction.
• Propose and evaluate alternatives: AI technologies can be used to propose and evaluate alternatives to problems caused by product aging. Taking into account ontological knowledge and constraints, AI models can analyze the characteristics, performance, and cost of similar products and materials, and provide a choice of alternatives to help solve problems caused by aging or upgrade products.

The use of AI technology will enhance decision support in product aging management combined with ontology-based knowledge representation, allowing AI to analyze data, predict, optimize, and suggest alternatives for quick and effective decision making.