Beyond Spreadsheets: Revolutionizing BOMs through Collaborative Intelligence

For years, Microsoft Excel has been the go-to software for Bill of Materials (BOM) management in many industries, particularly in manufacturing and automotive sectors. Its familiarity, flexibility, and accessibility have made it a staple tool for creating and maintaining BOMs. However, as supply chains become more complex and data-driven decision-making becomes crucial, the limitations of spreadsheet-based BOM management are becoming increasingly apparent.

01. The Excel Era: A User's Perspective

Before we start to improve the process we need to understand the current User journey. 35% of the small and medium customers are still use this file format.

1. Initial Adoption: Users start with Excel due to its ubiquity and familiarity. Many professionals already have basic Excel skills, making it an easy entry point for BOM management.

2. Basic BOM Creation: Users create simple BOMs, listing components, quantities, and basic attributes. Excel's flexible grid layout allows for intuitive data entry and organization.

3. Data Manipulation: As users become more proficient, they utilize Excel's sorting and filtering capabilities to organize BOM data. Basic formulas are employed for calculations like total costs or weight summations.

4. Customization: Users create custom formats, color-coding, and conditional formatting to enhance readability and highlight important information.

5. Sharing and Collaboration: BOMs are shared via email or network drives. Collaboration occurs through sequential editing or manual merging of changes from different users.

6. Version Control ChallengesAs product complexity increases, users struggle with version control. Multiple copies of the same BOM circulate, leading to confusion and potential errors.

7. Data Volume Limitations: Large BOMs with thousands of components start to strain Excel's performance, causing slow load times and calculation issues.

8. Integration Hurdles: Users find it difficult to integrate BOM data with other systems like ERP or PLM, often resorting to manual data entry or complex import/export processes.

9. Error Prone: Manual data entry and formula errors become more frequent as BOMs grow in complexity, leading to costly mistakes in procurement or production.

10. Scalability Issues: As the organization grows, Excel-based BOM management becomes increasingly cumbersome, lacking the scalability needed for enterprise-level operations.

11. Recognizing Limitations: Users and management begin to recognize the limitations of Excel for BOM management, particularly in terms of collaboration, accuracy, and scalability.

02. Transition to Digital BOM Solutions

The most manufacturer are using individual digital BOM solutions already.

01. Discovering New Capabilities: Users begin to explore advanced features like real-time collaboration, automated version control, and integration with other enterprise systems.

02. Adoption of Analytics: With data now centralized and standardized, users start leveraging built-in analytics tools for insights into cost optimization, supplier performance, and risk management.

03. Collaborative Workflows: Teams adapt to new collaborative workflows inside the company, with multiple users able to work on the same BOM simultaneously, enhancing productivity and reducing errors.

04. Integration Hurtles: Limited integration with other BOM solutions. A proper Data Integration could reduce manual data entry and improving overall data accuracy. Here a list of most used BOM solutions for manufacturer:

• Bosch Digital Twin - IAPM (Integrated Asset Performance Management)
• SAP Digital Twin Network
• SAP Business One Bill of Materials
• SAP Enterprise Product Development
• Siemens Teamcenter BOM Management
• OpenBOM
• Arena PLM
• NetSuite
• Fishbowl
• Autodesk Fusion 360 Manage with Upchain

05. Continuous Improvement: As users become more proficient with the digital BOM solution, they continually discover new ways to optimize processes and leverage data for decision-making.

06. Advanced data analytics: BOM are limited to run advanced data analytics.

03. Integrating BOM Data into Cloud-Based Data Lakehouses

BOM data can be integrated into cloud-based data lakehouses for advanced analytics, particularly focusing on time series analysis. This approach combines several modern data management and analysis concepts.

Data Lakehouses, such as those offered by Snowflake, Databricks, Microsoft Fabric, AWS RedShift, and Google Cloud BigQuery, provide an ideal environment for storing and analyzing BOM data alongside other relevant datasets. Here's how this integration steps can work:

1. Data Ingestion: BOM data, typically structured information about product components, can be ingested into the data lakehouse along with related time-stamped data (e.g., production logs, supply chain events, quality control data).
2. Data Storage: The lakehouse architecture allows for storage of both structured (BOM data) and unstructured or semi-structured data (e.g., sensor readings, maintenance logs) in a single repository.
3. Data Cataloging: Metadata about the BOM and related datasets can be managed through the lakehouse's catalog layer, making it easier to discover and use the data for analysis.

04. Time Series Analysis of BOM Data

Once the BOM data is in the lakehouse, you can perform advanced time series analysis:

1. Component Trend Analysis: Track changes in component usage over time.Identify trends in component substitutions or upgrades.
2. Supply Chain Optimization: Analyze lead times and availability of components over time.Forecast potential supply chain disruptions based on historical patterns.
3. Quality Control: Correlate component changes with product quality metrics over time.Identify patterns that might indicate quality issues related to specific components or suppliers.
4. Cost Analysis: Track component cost fluctuations over time.Forecast future costs based on historical trends and market data.
5. Product Lifecycle Analysis: Analyze how BOMs evolve throughout a product's lifecycle.Identify patterns in component obsolescence and replacement.

To implement these analyses, you can leverage the tools and capabilities provided by the data lakehouse platform:

1. SQL and DataFrame Operations: Use SQL or DataFrame APIs (e.g., Spark DataFrames) for data manipulation and basic time-based aggregations.
2. Time Series Libraries: Utilize libraries like Prophet, or statsmodels for Python to perform advanced time series forecasting and analysis.
3. Machine Learning Integration: Leverage the lakehouse's ML capabilities to build predictive models based on BOM and time series data.
4. Visualization Tools: Use integrated or connected BI tools to create dashboards and visualizations like Grafana, Kibana, Prometheus of time-based BOM analytics.

05. The Raise of Digital Twins

Many automotive manufacturers (OEMs) are already using digital twins for BOMs to optimize their value chains over the whole Lifecycle, also called X-BOMs. This approach brings several advantages.

• Digital twins in production lines and supply chains are continuously updated with real-time data from IoT sensors.
• Time series data from production machines, inventory, and deliveries are analyzed using LSTM networks to capture complex temporal dependencies and make precise predictions.

AI-Driven Optimization and Risk Management

• AI algorithms identify patterns and anomalies in time series data to detect potential bottlenecks or quality issues early on.
• Production processes can be dynamically optimized, increasing overall efficiency by 15%.
• Maintenance work is planned more precisely, reducing unplanned downtime by 25%.
• Supply chain risks are assessed in real-time, and alternative strategies are simulated.

Efficiency and Cost Savings

• Energy consumption is reduced by 10% through optimized process control.
• More accurate demand forecasts and optimized order quantities lead to a 15% reduction in inventory levels.
• Potential bottlenecks in the supply chain are detected early, and alternative suppliers are identified, minimizing costly production interruptions.
• Overall costs were reduced by 4.4%, and lead times were shortened by 24% through intelligent component selection and supplier optimization.

06. Standardization and Semantic Models

In this context, Tractus-X KITs (Knowledge Integration Toolkits) play a crucial role in enabling the secure and standardized exchange of data, particularly for BOMs in the automotive industry.

1. Standardized Data Models: The IDTA define digital twin sub-models as Aspect Meta Models (SAMM) that defines consistency and interoperability across different systems and organizations. Catena-X developed specific sub-models for the automotive: BOM as Specified, Part as Specified, Material for Homologation, ECLASS Dictionary
2. Secure Data Exchange Protocols: The IDSA facilitating the safe sharing of sensitive BOM information among partners. Dataspace Protocol for data contract negotiation and Digital Claims Protocol for Identity flow between
3. Integration Capabilities: The Catena-X standards allowing seamless connection between various data platforms and data spaces for various use cases.

07. Radical Collaboration in the Automotive Dataspace

Catena-X started an open-source project with Eclipse Tractus-X to support developers in order to accelerate the development of services and applications to contribute significantly to the rapid scaling of the Catena-X ecosystem.

• Network Services: provides the foundation for creating a data chain based on vehicles, parts, and materials: Industry Core KIT, Digital Twin KIT, Data Chain KIT, Connector KIT
• Product Life Cycle Management and Quality: Tracking components and materials throughout the supply chain with greater accuracy with prediction: Traceability KIT, Quality KIT
• Supply Chain Resiliency: Demand and Capacity KIT, PURIS KIT
• Sustainability Initiatives: Supporting the tracking and reporting of sustainability metrics for components and materials and visualize the data as Digital Product Pass and the Battery Pass: Circularity KIT, ECO Pass KIT, PCF Exchange KIT (Product Carbon Footprint)