The standard (improved video + element definitions) established through STEP 5 (selection → extraction → classification → improvement → verification) serves as the reference. This feature uses AI to view and align the repeated work videos of the target person (junior staff, new employees, etc.), then visualizes cycle time comparisons, delay per element, and AI coaching.

This allows you to see at a glance who is taking time on which element and where variation exists, making the priority of training and on-the-job training clear.

This article is STEP 6 of the Work Analysis series. For the full overview, see Work Analysis Overview and Getting Started.

Preparation: Confirm the Standard

Before running a comparison, you must complete STEP 1 through 5 (selection through verification) to confirm the standard video and element definitions. The comparison step cannot be run if the standard has not been established.

Add Comparison (Choose Target Person and Video)

1. Open "Compare" (STEP 6) from the rail on the left side
2. Click [Add] in the top right
3. In the dialog that appears, configure the following:
   • Target person: Select from team members (records whose work video)
   • Work video: Select the target person's repeated work video from [Materials]
4. Click [Start] to begin AI video analysis (you will be notified by email when complete)

Comparison List and Details

The comparisons you add appear in the Members list. You can see the target person, video, analysis date, and status. Click a row to open the detailed screen (comparison layout).

• While analyzing, "Analyzing" is displayed; for partial failure, "Partially completed"; for failure, "Failed"
• Completed comparisons show no status display (only states requiring attention are highlighted)

Comparison Layout (View Standard and Target Person Side by Side)

In the detailed screen, the standard video appears on the left and the target person's video on the right, with playback position synchronized between both. In the comparison timeline at the bottom, the target person's delay is visualized with red shading.

Cycle time (Stacked Bar Chart)

• Left side = standard cycle time, followed by each cycle for the target person
• Each bar displays elements stacked = cycle time total
• Shows reference lines for standard (blue dashed line) and target person average (orange dashed line)
• Click a cycle to switch to detailed comparison for that cycle

Delay per element

Shows the difference from the standard (set to 0) per element in a line chart.

• Selected cycle: Solid line (orange) + red shading
• Average: Dotted line (light color)

Variance per element (For 2 or More Cycles)

For each element, you can compare the target person's shortest – longest range, average, selected cycle, and standard on one screen. You can see at a glance whether the element is produced consistently or has large variation.

AI Coaching (Improvement Suggestions)

From the [Improvement suggestion] tab in the right panel, AI can generate personalized coaching for this target person.

1. Click [Generate suggestion] in the top right
2. AI generates improvement points at the element level from the differences between the standard and target person
3. Each suggestion displays importance (high / medium / low), the target element, point, and advice

If you're not satisfied, you can try again with [Regenerate suggestion].

Change Target Person / Adjust Cycle

• From the top of the detailed screen, you can change the target person later (if you need to correct whose data it is)
• From [Cycle adjustment] in the top right, you can fine-tune if the AI's automatic cycle boundaries are misaligned for the target person
• If you're not satisfied with the analysis results, you can try again with [Re-analyze]

Fix Target Person's Measurements by Hand

In [Cycle adjustment], in addition to fine-tuning boundaries, you can perform the following operations.

• Split element / Merge with next / Return to not detected
• Add measurements to elements that were not detected (add from the gray chip)
• Undo / Redo
• Warnings for extremely long or short elements

Previously, if AI missed an element, that element would remain blank. Now you can add measurements by specifying the interval while watching the video.

The "Match with the reference" bar shows which references exist in that instance and which do not. This is information that cannot be drawn on the time axis Gantt chart, so please check it together.

Which Standard Was Used for Comparison

Comparison results are based on the standard that AI used for matching. The numbers, the bar on the left side, and the video all use that standard. Even if you add, remove, or rearrange elements on the standard side later, the meaning of results already displayed does not change.

If you edit the standard later, the detailed screen top will display "results compared against which standard". When you want to review with the latest standard, click [Re-analyze] at the top to redo the comparison.

For comparisons performed before September 13, 2026

Comparisons before that time do not have a record of the video position on the standard side. Therefore, the bar and video on the standard side display the current standard (the numbers use the standard at the time of matching as before). If you edit the standard later, the basis will be split, which we will display on the screen. Redoing with [Re-analyze] will then display the standard side as it was at matching time.

Handling Failures and Partial Failures

• Failed: AI observation and error details are displayed. After confirming the cause, execute [Re-analyze]
• Partially completed: When there are failed segments, you can re-run just that section with [Re-analyze failed chunks only] (does not consume AI count)

Tips for Effective Use

• Polish the standard before comparing: If the element definitions on the standard are rough, the comparison results will also be rough. The more carefully you create the model in STEP 1 through 5, the higher the accuracy of the comparison
• Prepare multiple cycle videos: When you prepare 3 to 5 or more cycles per analysis of the target person's repeated video, variation and averages come out more accurately
• Use as OJT training material: The specificity of "You are delayed by X seconds on element Y; this is how to move faster" greatly changes the conversation on the shop floor
• Re-run comparisons regularly: As a way to measure training effectiveness, re-running comparisons of the same task monthly or quarterly can visualize improvement

Related

• Work Analysis Overview and Getting Started
• STEP 5 Verify: Compare Before and After Improvement and Export Video