Abstract
Traditionally, the transition from design stage to work order generation suffers from lengthy cycle times due to insufficient requirement analysis and unforeseen challenges. This paper proposes a novel concept leveraging Model-Based Systems Engineering (MBSE) and AI-powered predictive modeling to significantly reduce this cycle time. Our approach comprises three key stages:
- Structured Requirement Analysis: In-depth interviews with designers and domain experts reveal inconsistencies and variations in product and process parameters. This unstructured information is converted to structured data with assigned values, facilitating analysis and prediction.
- Functional and Risk Assessment: The converted data is utilized to conduct a comprehensive review of integration, development phases, functional analysis, risk identification, supply chain issues, cost analysis, and feasibility studies. This generates valuable insights for feedback and optimization.
- Predictive Modeling for Work Order Generation: Based on historical data and insights from the assessment stage, a predictive model estimates the change in time required to generate work orders due to specific design modifications. This enables proactive adjustments and avoids delays caused by missing information or material unavailability. The proposed approach addresses critical challenges:
- Improved Communication: MBSE provides a centralized platform for seamless information exchange between designers, engineers, and domain experts.
- Early Fault Detection: The functional and risk assessment stage identifies potential issues early, allowing for corrective action before work orders are issued.
- Reduced Time-to-Production: AI-powered predictive modeling minimizes design-to-production delays, enhancing efficiency and competitiveness. By combining MBSE and AI, this paper presents a novel and promising framework for drastically reducing cycle times, leading to significant cost savings and improved product launch.
Speaker
Sanghamitra Goswami
currently works in Blue Origin. Before joining Blue Origin, she worked in retail, technology, and financial services industries, using data science and machine learning to solve problems related to demand forecasting, real time monitoring of inventory, fraud detection and customer retention. She did her graduate and undergraduate education from India, completing Ph.D. from Indian Institute of Science, Bangalore and subsequently a post-doctoral fellowship from Kellogg School of Management, Northwestern University. In addition to her work, Sanghamitra is passionate about providing access to aerospace and STEM education to underserved communities. In her free time, she conducts online STEM education programs for underserved community schools across India, igniting a love for science and technology in young minds. She is also interested in arts and dreams of running her own studio, where people can participate in creating music and art amidst their busy professional life.
