Batch scheduling in a lean manufacturing world

Mike Liddell

The concept of creating batches in the manufacturing process is prevalent in many industries. In fact, economies of scale demand it. An example of this would be a paper mill needing to mix a vat of a specific pulp from which orders belonging to a certain product group can be made. It is not economically feasible to mix a vat for each customer’s order.

Good news and bad news arises for manufacturers using batching. The good news is that as long as they can survive using buffers and manufacturing what they need to replenish forecast demand, then a standard off-the-shelf manufacturing resource planning (MRP) system will address their needs.

The bad news is that to reduce waste and move toward a demand-driven, lean manufacturing model, they will need a completely different tool set. Otherwise, things will become unmanageable.

Still, the best news is that systems are specifically designed to address the needs of those manufacturers who must manufacture-to-order while continuing to run batches.

The Problem
Lean manufacturing means eliminating waste, and one way to quickly eliminate waste is to move toward a make-to-order, demand-driven business model.

The price MRP systems pay when grouping like-orders together is that they create work orders that cannot be easily tied to actual customer demand. This is not a problem in a make-to-stock world, but it is a major headache in the make-to-order world.

The key component needed in the make-to-order world is a scheduling system that schedules and tracks production at the sales-order level. This is not really an issue in machine shops where each operation of each order is scheduled independently.

A simplified way to think of batch scheduling is to compare it to the way airlines schedule passengers. The batch is the flight, and the number of seats on the plane limits the size of that batch. Passengers are the orders, and the time that each passenger can schedule the next phase of his/her journey is controlled by the time the flight is scheduled to arrive at its destination.

The Solution
A Batch Scheduling System is a solution that provides four essential capabilities:
1. Create and schedule batches
2. Assign orders to batches
3. Edit the batch contents
4. Schedule additional operations.

Create and schedule batches:
The scheduler needs to predetermine the type and size of the batches and use the scheduling system to arrange the dates for the batches to run.

This can be done based on company policy, experience or forecast. In some situations, the system may need to create new batches to satisfy orders that cannot fit into an existing batch.

If four empty batches are created, each with the same capacity and an expected duration of three hours, the schedule looks like this:

Batch Scheduling empty


Assign orders to batches:

When new orders are introduced into the scheduling system, there are three major rules that determine if these orders can fit into a batch:

Batch Capacity Rule: Most batches have a finite capacity. This rule determines whether enough capacity is available for an order. Capacity can be expressed in units, such as pieces, tons or hours.

Date Rule: This rule uses the date of the batch and the customer request date to determine the best batch to use.

Type Rules: Type rules compare one or more attributes of the order with one or more batch attributes to determine if the order can fit in a batch. This rule can be applied to anything from matching colors, sizes or raw materials, to applying a certain finish or routing step.

As orders are assigned to batches, the schedule will effectively look like this.

Batch Scheduling

Some manufacturers, such as those making high-pressure laminates, have an additional constraint. One batch may actually consist of a fixed number of books, so the system must be able to handle an additional layer of rules that controls the capacity at the book level and what orders can and cannot go into the same book.

Edit the batch contents:
Since a batch appears as just one line on the Gantt chart, the scheduler needs a tool that provides views of both the orders in a batch and batch specific data. This tool must allow the scheduler to drag orders in and out of batches without breaking any of the rules. The purpose of this tool is to enable fine-tuning. This tool will need the following capabilities:

  • Drag and drop orders between batches.
  • Allow scheduler to override rules.
  • Pull order from unscheduled box to batch.
  • Split orders to spread over more than one batch.
  • Create remake orders.
  • Modify attributes of both orders and batches.

In the following software graphic, by clicking on the resource in the left-hand column, the Batch Editor shows the data for each batch such as the Steel Group and the Allocated Capacity.

Batch Editor

By clicking on a batch, a scheduler can see and modify order level data.

Modify Order Level Data

The Un-allocated Orders box shows the orders that have not yet been allocated to a batch. These can be sorted and dragged to a batch.

Un-allocated Orders

Schedule additional operations:
In many cases, the scheduler must assign orders to additional operations such as finishing or packaging. This necessitates setting the start date of the finishing operation to the end date of the batch and then scheduling it. This allows the scheduling system to calculate completion date for each order. The Gantt chart below shows how the finishing operations for each order are tied to the time its batch completes at the furnace.

Gantt Chart

Benefits
Such a system has some very attractive additional benefits:

  • It reduces non-value-added-time on the front end of an order by enabling manufacturers to quickly provide clients with an accurate promise date.

  • Because batch scheduling works within a predetermined framework, it is less likely to react unpredictably to cascading changes.  

  • Most of the number crunching is done by the system, so the scheduler has more time to evaluate and fine-tune the schedule.

  • It is easy to track the completion date of orders.

  • Additional functionality is available to let the scheduler make valid substitutes when faced with material shortages.

  • It provides a tool for capacity-constrained companies to synchronize schedules with business objectives by creating batches for high-margin products rather than promising delivery based on first come, first served.

Summary
As technology changes, the world looks toward customization. This dooms products to shorter life spans, and it becomes difficult to accurately forecast demand. Additionally, companies in Europe and North America consistently compete with the low-cost mass production coming from places like China. By retooling companies to react faster to customer demand, the leaders will move away from a “sell-what-you-make” business model to a “make-what-you-sell” business model.

New systems like the one described here provide a way for companies to impact their bottom line in a radical way that will give them a significant competitive advantage in their market.

Listed below are some industries that currently use this approach:

  • High-Pressure Laminates
  • Paper Mills
  • Steel Mills

Other likely candidates include:

  • Chemicals
  • Food
  • Glass
  • Plastics

About the author:
Author Mike Liddell is the chief executive officer of Suncoast Technology Partners based in Bradenton, Fla. For more information about Mike or this company, visit www.stpartners.net.

Subscribe to Machinery Lubrication

About the Author