In Parts 1-10 of this blog series, we built a digital thread for an autonomous vehicle system to demonstrate how a federation of models in different software tools can become a unified specification of the system. Parts 11-13 demonstrated how Syndeia Web Browser’s standard reports and visualizations allowed a user to access that information. In Parts 14 and 15, we started to build a custom reporting capability using direct call to the Syndeia Cloud REST API, initially through a Jupyter Notebook and then through a digital pipeline built in the Airflow platform.

The final piece of the process is to create a custom dashboard where regular users can view both current data and historical data from previous digital pipeline runs. Key desirables are

• The Dashboard Development Environment (DDE) should be a familiar platform for data scientists and application developers, rather than uniquely Syndeia-specific. The wide availability of well-developed DDEs reduces the learning curve for digital thread implementation.
• The DDE should have an extensive library of widgets for reading and writing data, creating tables, graphs and other visualizations, and interfacing to other applications (e.g. publishing, LLMs), to support rapid deployment of low-code applications.
• The DDE should be able to adapt the code used earlier in the custom notebook and digital pipeline with minimal changes.

Figure 1 Python Dashboard Code in Streamlit UI 

Streamlit is an open-source Python framework for data scientists and AI/ML engineers to deliver interactive data apps in only a few lines of code. Figure 1 shows a screenshot from the Streamlit DDE for the Calculate_Project_Metrics section of the Python code. For comparison, check Figure 4 in Part 14 from the Jupyter notebook and Figure 2 in Part 15 for the Airflow digital pipeline to see that the same Python code has been used in three different places. This code extracts the digital thread data from Syndeia and the underlying repositories and calculates the metric values. The only major additions call the Streamlit API for accessing stored data and reporting.

Figure 2 Jama-SysML Dashboard (top), Autonomous Vehicle Project

A user logging into the data app created on their web browser sees a variety of relevant data. Figure 2 shows the top part of the custom dashboard, comprised of real-time data about the Jama requirements that are part of the AVDZ01 digital thread project generated from queries to Syndeia as the dashboard is displayed. The table has three tabs

• All Connected Jama Artifacts, which shows all Jama artifacts (requirements, in this thread) connected to SysML artifacts
• Added Last Seven Days, which shows newly connected Jama artifacts within the last seven days, i.e. recent activity in this section of the digital thread, and
• Update Required, which shows all connected Jama artifacts that have been changed since the connection itself was last updated. Such relations warrant individual examination to see if the change impacts the downstream SysML artifact.

The two pie charts display two key attributes of the Jama artifacts, Status and Priority, and change depending on which of the three tables above is selected. Note once again that these attributes are read real-time directly from the Jama repository; they are not normally stored in the Syndeia graph database.

Figure 3 Jama-SysML Dashboard (bottom), Autonomous Vehicle Project

Figure 3, from the bottom half of the dashboard, shows historical data from digital pipeline runs for the Jama-SysML metrics for the autonomous vehicle project. This data is stored within the Airflow platform as an XCom in a standard Python dictionary and read by Streamlit when the dashboard is generated. The Activity metric, for example, shows a drop-off in the activity of adding new Jama artifacts to the project in early August, with minimal additions since that time. However, the Completion metric indicates that nearly all the candidate Jama elements are already linked, i.e. completion approaches 100% (1.0). Historical data is particularly useful for metrics where the absolute value may be less important than the trends over time.

In parts 11-16 of this series, we have revisited the autonomous vehicle digital thread created earlier, exploiting the latest features of Syndeia for both vendor-supplied and custom user interfaces. In particular, we have demonstrated how relatively simple Python code calling the Syndeia REST API endpoints could be used successively in Jupyter notebooks for development and validation, Airflow digital pipelines for workflow automation, and Streamlit dashboards for project management.

We plan to add to this series as new capabilities become reality. Two frontiers of digital thread technology are digital twins and agentic AI. While digital threads of the type that Syndeia builds are not digital twins themselves, i.e. orchestrated simulations emulating system performance, they can support digital twin work by configuration-managing the combination of data elements to be simulated. Agentic AI is already being demonstrated for accepting natural language queries within Syndeia-based project dashboards, but using it to help build the digital thread is an active research area. Keep a watch on our blogs for the latest developments, or contact us at info@intercax.com to discuss your own use cases.

In the following blog posts in this series, we will demonstrate some of these current capabilities and offer a few glimpses of the future

• Part 11 - Model-Based Systems Engineering for Autonomous Vehicles, Part 11 – Digital Threads 
• Part 12 – An introduction to Digital Thread Projects, Relations, Baselines and the Syndeia Web Dashboard 
• Part 13 – Reports, Graph Analysis and the Digital Thread Explorer
• Part 14 – Open REST API for scripts, Jupyter notebooks and project metrics 
• Part 15 – Digital Pipelines 
• Part 16 – Custom Dashboards (This Part)
• Part 17 – Digital Threads for Agentic AI (To be released)

Missed the earlier parts? Check them out here.

To evaluate Syndeia with your own toolset, or just to discuss your requirements and use cases, send your questions and requests to www.intercax.com/help and let us help you adopt best practices in Digital Engineering.