Line balancing used to depend on experience and personal judgment. Now the system calculates the best balance point for you based on actual data, in seconds.

AI-Powered Production Line Balancing

Finding the right number of machines and workstations to hit a specific production target took time, relied on guesswork, and left room for costly mistakes before the line even started.

That changes now.

We've launched the AI-powered production line balancing feature. Configure a few settings around your order and your line, and the system proposes the optimal workstation and machine distribution to hit your target in seconds, before you start production.

To access: open the style page and you'll find a new field called Production Line Balancing. Press it to enter.

Production Line Balancing

How It Works

1. Basic Settings

Start by entering the core data for your order:

• Order quantity you want to balance the line for.
• Number of production days available and working hours per day.
• Time basis: either the style's standard SAM or efficiency rate.
  

Basic Settings

2. Stage Group Selection and Output Range

• Select the stage group you want to balance.
• Set the minimum and maximum output range, the window within which the AI will suggest a production quantity.
  

Stage Group Selection and Output Range

3. Advanced Settings

• Output step: the increment by which the system increases the proposed output in each step. Example: if set to 5, the system adds 5 pieces per proposal.
• Maximum operations per station: the maximum number of stages that can run together at one workstation. If more than one, the system automatically groups stages from the same stage group.
• Maximum machines: sets the upper limit on machines the AI can propose. Leave it empty to let the system work without a cap. Note: the maximum cannot be set lower than the number of machines the system has already proposed.
• Capacity allowance: a percentage that determines when the system proposes adding a new machine. Example: with 10% allowance and a target of 100 pieces per hour, the system allows load up to 110 pieces — anything above that triggers a machine addition proposal.
• Exclude manual operations: the AI automatically removes all operations classified as manual before running balancing calculations.
  

Advanced Settings

4. Run the Balance and Read the Results

Press Balance Production Line and within seconds the system displays:

• Suggested hourly output: the optimal balance point for the line, expected daily output: based on the working hours you defined, expected completion: across the number of days specified.
• Suggested number of machines and workstations.
• Wasted capacity: a percentage representing time that won't be utilized in production.
• Bottleneck station: the station creating the main constraint in the line.
• Suggested production multiples: if you plan to run more than one line on the same quantity.
  

AI Balance Results

5. Yamazumi Chart

Displays the actual cycle time per station versus takt time, with filtering available by stage group.

Yamazumi Chart

Balancing Versions & Print Preview

6. Station Layout Table

A table showing each stage group and the stages under it. For each stage:

• Standard SAM, Suggested SAM.
• Suggested number of machines.
• Waste percentage — how much of the available time this stage is actually utilizing.
• Dependencies column and flags column.
  

Table view

The layout also shows how the production line should be arranged according to the AI's proposed distribution, helping improve workflow and achieve the best output allocation.

Layout view

Manual editing

7. Machine Summary

Displays the total number of machines required per machine type to achieve line balance, based on the inputs and settings you've configured.

Machine Summary

Frequently Asked Questions

• Can I balance only part of the style's stages?
  Yes, you can select a specific stage group without applying balancing to the entire style.
• What does leaving the maximum machines field empty mean?
  It gives the AI full freedom to propose the optimal number of machines without any cap.
• What's the difference between standard SAM and suggested SAM?
  Standard SAM is the pre-defined time for a stage, while suggested SAM is what the AI recommends to achieve optimal balance.
• What is the bottleneck station and how do I use that information?
  It's the station creating the main constraint slowing down the line. Identifying it lets you focus on resolving it first to improve overall line efficiency.

Have a question about this update?

Reach out to the Garment IO team. We're here to make sure you're getting the most out of every feature.