Unit Flow (Beta)

The new Unit Flow module is a powerful tool designed to provide comprehensive visibility into the flow of units across the production line. This feature offers an adaptable overview, allowing users to visualize and analyze production data from various perspectives. 

The flow is mapped based on test and repair reports, generating a main process flow while highlighting units that deviate from the standard path. This includes units that bypass certain steps, follow an unusual order, or skip processes altogether.

By leveraging this tool, users can quickly identify root causes of production issues, optimize process efficiency, reduce operational costs, and take proactive steps to prevent potential disruptions. 

This feature is currently in beta and available exclusively to select customers. For early access, please see WATS Unit Flow BETA Early Access Program.

Accessing the Unit Flow module

The Unit Flow module can be accessed in two ways:

1. Via the Analytics navigation menu: Navigate to the Analytics section in the main menu, then locate the Manufacturing Insight area. Under the Flow view section, select the Unit Flow option.
   

2. Drill down from the Process Heatmap: From the Process Heatmap module, select a row and click on the Unit Flow icon to view detailed flow data for the selected units.

Filter

The Unit flow filter in the toolbar allows users to filter reports by specific criteria such as serial number, part number, level, process, and run within a defined timespan. Once applied, the filter displays the entire flow for the units matching these criteria. The max result setting limits the number of unique units the filter will return, displaying up to the set maximum of recent reports.

Flow tracking

The module sequences each unit's progress across processes, consolidating retests and providing a detailed view of each unit’s journey.

• Start: Displays the number of unique units entering a process.
• End: Shows unique units existing in the process. This may remain blank if data is incomplete.
• Both Start and End counts reflect unique units, regardless of the selected view and option

This accurate tracking approach helps users identify incomplete or irregular flows within production.

Views

Historical 
The "Historical" displays all units that have been processed, including retests.

• Users can analyze the flow based on either Reports or Units (unique units), allowing for flexible and precise evaluation of production data. In the Current View, the count is set to Units by default, ensuring consistent real-time and temporal data tracking. Meanwhile the Time View, is set to Reports by default. 

This view provides several options for drilling down into station-level details: 

• Double-click the link label or node to drill down, or 
• Use the "Split by stations" option from the drill-down menu. 

There are two ways to split by stations: 

• Split by station from a link: This splits the target and source process data into individual stations based on the link data. 
• Split by station from a node: This shows all units and processes associated with the selected process and splits the selected process into stations, while keeping other related processes unsplit.  

Additionally, the top three streams will feature a wider link body, visually emphasizing their significance within the flow. 

Note: Splitting into stations is only available in the "Historical" view. 

Current 
The "Current" view shows the current location of units within the production line and predicts their next steps based on the usual flow. This helps track where each unit is in real-time and anticipate the upcoming processes. 

Time  
The "Time" view provides an overview of the execution times: 

• It shows the time taken from start to end, within a process, and between processes. 
• The default time displayed on link labels and nodes is the median time, but this can be adjusted to show min, max, or average times using the time options. 
• Hovering over a link label or node will display a tooltip with the min, max, average, and median times. 

Time calculation: The time spent in a process and in transit is calculated per sequence. If a unit enters a process multiple times, individual time spans are calculated for each entry, and the median time is derived from those sequences. This approach ensures more accurate time data than simply using the total span between the first and last entry of the unit into the process. 

Link types

The Unit Flow module uses several link types to represent different unit movements within and outside the main production flow. Here are the default link types: 

• Main Flow: Represents the primary flow of units through the standard production process. 
• Outside Main Flow: Indicates units entering or exiting external processes that are outside the main flow. 
• Non-pass: Marks units transferred to the next process without passing the required criteria. 
• Repair In: Units that are sent for a repair or maintenance process. 
• Repair Out: Units that have been successfully repaired are re-entering the flow. 
• Jumped: Indicates units that bypassed one or more processes. 
• Wrong Order: Identifies units that have been sent back to a previous process, deviating from the correct sequence. 
• Retest: Represents units that have undergone retesting within a station or process. 

Additional Link Types by View: 

Current view: 

• Passed units: Indicates unique units that have successfully passed the current process. 

These link types provide clear visibility into how units are progressing, highlighting deviations, and ensuring easier analysis of production flows. 

Tracking units by serial number

When you click on missing units, units currently in process, or retested units in the graph, the system will retrieve the corresponding serial number. The serial number will be displayed on the right side of the screen within the graph settings.

Regardless of the selected view or option, the feature always returns unique serial numbers for consistency and accuracy.

By clicking on the serial number, the graph will highlight all nodes and links representing the complete journey that this specific unit has gone through in the production process. This allows you to easily trace the path of the unit and analyze its flow in detail. 

Additionally, serial numbers can be copied for further analysis or used in other reports to investigate specific units more deeply. This feature provides a clear and actionable way to track individual units across processes. 

Analyze data 

You can easily analyze data by interacting with the graph in two ways: 

1. Clicking on a link label, or 
2. Clicking on the menu icon in a node to open a dropdown menu with analysis options. 

From the dropdown menu, you can: 

• Retrieve Units in the process, Retested units or Missing units. 
• Use the "Analyze with units in process" option, which takes the initial data and resubmits it to the filter. This generates a new graph, allowing you to further analyze the data based on the units currently in the process. 

These options offer a flexible and intuitive way to drill down into specific data points and analyze unit flows with ease. 

Retest tracking

Retest Tracking Behavior

• In the Reports option of the Historical view, retest counts each subsequent test in the same process after the first, each time the unit enters the process.
• In the Units option of the Historical view, retest counts each unit that has been tested multiple times in a row in the same process.
• The total number of retests is determined by the total count of all retest links.

Process-Level vs. Station-Level Retest Counts

The retest count differs between process-level and station-level views to provide more granular tracking:

• Process-Level (Historical View): Retests are counted at the process level.
  • If a unit is tested in process A, then B, then A, that does not count as retest.
  • If a unit is tested in process A, then A, then B, then A, then A, then A, that counts as 3 retests in process A.
    • With the Units option it is counted as “1 unit has been retested”.
• Station-Level (Split-by-Station View): Retests are counted separately at each station, offering a more detailed view of retest activity per station within a process.
  • Within the same process, if the unit is tested on station A, then B, then A, that does not count as retest, even though in Process-level view it was retested twice.
  • Within the same process, if the unit is tested on station A, then A, then B, then A, then A, then A, that counts as 3 retests on station A; but counts as 5 retests in Process-level view.

This approach supports a comprehensive analysis of unit performance across both processes and individual stations, enabling users to assess retests accurately at multiple levels.

Customizing and Exporting the Unit Flow Graph 

You can customize and interact with the Unit flow graph through various settings and options: 

1. Graph direction: 
   Set the graph direction to your preference using the "Graph options" located at the top-right corner of the graph. 
2. Exporting the graph: 
   The graph can be easily exported as a PNG file, making it convenient for sharing or reporting. 
3. Link options: 
   Located on the right side of the graph, link options help manage the display of links. This tool is essential for reducing visual noise and focusing on relevant data by offering the ability to: Hide, highlight, or remove links that represent fewer units than a specified threshold. 
4. When you hover over a link in the link options, it will be temporarily highlighted. 
5. Clicking on a link name will lock the highlight until clicked again or another link is selected. 
6. To hide a link, simply check the corresponding checkbox. 
7. Graph navigation:
   For easier navigation and viewing, use the navigation bar at the bottom-left of the graph to: Switch to full-screen mode. Fit the graph to the screen. Zoom in or out. Navigate using a mini map, allowing quick access to different areas of the graph. 

These features make it easy to tailor the graph to your specific needs, focus on key areas of interest, and export your findings for further analysis.