Finding the Common Denominator
One of the more common refrains heard from Job Shops is that "Lean Manufacturing obviously works in OEM (original equipment manufacturers) like Toyota or John Deere because they make pretty much the same thing everyday. But, we are different, we build on demand, we make anything and everything."
While this statement is absolutely true for the majority of make-to-order shops, there are enough similarities between OEM environments and Job Shop environments to negate any argument that Lean can be expected to work only in high volume-low mix settings.
The tipping point for many Job Shops is the realization that somewhere in the vast array of part numbers and product types meandering through their shops, there is a hidden set of common denominators: something similar about the products that can become a vehicle for bringing commonality to an otherwise uncommon mixed-model product line.
The application of a manufacturing rhythm (Takt Time) can only be possible if you can represent "demand" in a common form or unit of measure. Henry Ford could easily calculate a Takt Time by dividing the available time in an eight-hour day by the number of cars he wanted.
For example: a 480 minute day divided by a 240 car demand means that the manufacturing process must produce a car every 2 minutes. But what about a sheet metal shop that makes computer cabinets one day, hospital equipment the next day, and components for go-carts the following day? How can a rhythm be established when the parts vary in complexity, required machine time, or when they run at significantly different rates?
A manufacturer of oil field pipeline equipment in Edmonton Alberta Canada was working to establish their first model line (cellular manufacturing team). A PQR (product-quantity-routing) analysis determined that they would turn (lathe), machine, drill, and perform secondary operations on some 7000 units in the coming year.
Trouble was, of the 100's of potential part numbers that made up the 7000 unit demand, there was little similarity. Some parts had multiple features; others were simple pieces of pipe with a single flange. Some were the size of a bread box, others were machined from 600 pound blocks of forged material.
For this make-to-order shop the first step was to determine a reasonable Takt time. Calculating the available time was easy. 250 days per year, 8 hours per day, 60 minutes per hour: 120,000 available minutes per year. Takt time per unit was determined by dividing 120,000 minutes by 7000 units. This resulted in a Takt time of 17.1 minutes per unit.
But they soon realized that the variation in run speeds and process times, would make this simplistic way of looking at Takt time unrealistic. Seeking a truly rhythmic and harmonious manufacturing process that could be measured on a linearity (hour by hour) chart would require identifying a hybrid unit of measure and a hybrid Takt time.
When they examined the 7000 units required for the year, 2500 units were categorized as small and easy to handle. They decided to use this category of parts as a common denominator, equal to one unit of work. An additional 2500 were of moderate size and complexity and therefore of moderate difficulty. After performing time studies, this family of part types was found to be 1.6 times as difficult as the easy parts.
The remaining parts were large or multiple featured parts that required over twice as much labor and machine time. Additional time studies showed these parts to be 2.3 times as demanding as the easiest group of parts.
The next step was to perform a calculation to determine how many "units of work" were needed to produce the 7000 actual parts:
500 x 1 (easy) = 2500
2500 x 1.6 (moderate) = 4000
2000 x 2.3 (difficult) = 4600
The total number of work units required to manufacture 7000 parts was 11,100
Now the Takt time calculation can be changed to reflect the new (common denominator) demand of 11,100. 120,000 available minutes divided by 11,100 equals 10.8 minutes per work unit.
If the team is making the most difficult parts on a particular day, they will generate 44.4 work units. If the parts are easy they still generate 44.4 work units per day.
The number of actual parts will vary, but the number of work units will stay the same. Allowing a rhythm to develop. The team gets 2.3 credits for every difficult part they make. Only one credit when an easy part is produced.
Armed with this hybrid Takt time, the team could now effectively establish the ideal number of operators, balance their line, and set minimum levels of standard WIP (work in process inventory).
They reduced in-plant inventory by 78% (equal to over $200,000 of savings) and reduced distance traveled by 80% (saving 254 miles of material travel per year). Labor cost also went down because material handling is now done close to at arms reach rather than having to find a forklift, a place to store the materials, and then later perform the routine in reverse after locating the parts. A huge reduction in non-value-added activity now allows the team time to focus on even greater improvements.
This same technique has been used with shops who manufacture surgical tubing, computer touch screens, log chipping equipment, artificial hips and knees, and dozens of other products in a variety of companies and industries. This same process works in the administrative or engineering departments of your organization.
This technical information has been contributed by
ESPI - Enterprise Systems Partners, Inc.
Click on Company Name for a Detailed Profile
Home | About Us | Back To Technical Library | Contact Us Copyright © 1996-2010 JobShop.com. All Rights Reserved.
General or Technical Questions? E-mail support@JobShop.com