This article is based on Dr. Andersonís experience advising Hoffman and the article that he asked Hoffman to write for the Mass Customization issue he edited for the Agility Forumís Journal, Agility and Global Competition (Vol. 2, No. 2) Spring 1998. That article was written by Chris L. Conway, VP of Technology and Design, and Jerome M. Tiry, Manager of Engineering Services.

    Hoffman/Schroff, a division of Pentair Corporation, manufactures a vast variety sheet metal enclosures for electronics and electrical controls. The Hoffman catalog is almost two inches thick and contained over 6,000 different products and accessories.

    Customers ordered either "standard" products or custom products, which could only be built after all the materials and parts were ordered and received. The lead times were becoming unacceptable to customers and customized products aggravated this further.

    The variety of parts and materials caused a logistics nightmare. There were several hundred different blank sheet metal parts that had to be ordered in small quantities (often with significant setup charges at the supplier) with the customer waiting the whole time until all purchase orders were delivered. Since the sheet metal pieces were quite large -- up to 120 inches long -- raw material distribution consumed a significant amount of already-limited floor space in the factory. This situation was made worse considering that the sheet metal was on pallets which could not be stacked and needed fork lift aisles.

    To help get the project underway, consultant Dr. David M. Anderson was brought in to present his vision for Spontaneous Build-to-Order and Mass Customization, tailored to the manufacture of electrical enclosures. Through a series of presentations and meetings, he confirmed and reinforced Hoffmanís direction and offered advice and many ideas on implementation. He customized his general model into the a vision for agile sheet metal production


    Hoffman represents an ideal case for mass-customized, built-to-order products because of the high ratio of output variety (the products) to input variety (the raw material). Thousands of finished products can be made from a few types of sheet metal and other basic raw materials. Further, CNC (Computer Numerically Controlled) machine tools could be programmed to make parts efficiently in a "batch of one" mode.

    Hoffman manufacturing needed both Spontaneous Build-to-Order and Mass Customization:

1) Build-to-order a large variety of "standard" products, each in very low volume, often approaching batches of one.

2) Mass customize products to include special features, mostly custom holes and cut-outs for wiring and plumbing, and also such features as mounting holes, shelves, windows, vents, and color options.

    As with most mass customization, this activity represents an major opportunity for Hoffman to expand revenue and profits. It is very difficult for customers to add holes and cut-outs to a finished enclosure, but potentially easy for Hoffman if they can be cut by the programmable laser cutter that, in the mass customization model, also cuts out the blank from standard sheets. Given the high value to the customer and the relative ease of manufacture to Hoffman, this represents a high return opportunity while, at the same time, delighting customers, who are spared from slow and costly modification activities -- a real win-win situation! This opportunity is part of the rational justifying the mass customization project.

    Hoffman applied the following mass customization principles. First, they standardized raw materials to the point where a steady flow of those materials will come in to the plant and will be used one way of another.1 This eliminates ordering, and waiting for, raw materials. Second, provide the remaining variety easily with CNC machine tools in through agile manufacturing.2


    The biggest paradigm shift for Hoffman operations was the abandonment of the slow and costly practice of ordering sheet metal blanks. This would be replaced by cutting blanks to-order on programmable laser cutters from standardized sheets. If all sheet metal parts could be made from only one standard sheet size, material logistics would be the proverbial "no-brainer:" just arrange for a steady flow of standard sheet that matched the output of the plant.

    Some people, especially those firmly grounded in mass production and traditional value analysis, were initially uncomfortable with this. They were concerned about perceived waste of sheet metal if the required panels did not use 100% of the standard sheets. The two arguments that counter these fears are as follows: First, the cost of such waste would be very small compared to the material logistics savings from not having to order, transport, store, and distribute hundreds of different sheet metal blanks. Second, as the operation becomes more sophisticated, sheet metal part "nesting" efficiency will improve. In fact, sheet metal "scrap" that is not used in the enclosures themselves, could be used as source material for kanban part resupply (discussed later), where other parts, like stiffeners, lift angles, and brackets are made from blanks that are "squeezed in" between enclosure blanks.

    In Hoffmanís case, two thicknesses of plate were ordered, 10 gage and 12 gage. And each thickness was to be ordered in three sizes. In order to build products quickly to-order, rough forecasts are used to arrange for a steady flow of these six different sheets. But, Hoffman deems this feasible for this standardized set of raw materials, whereas it would been impossible to try to forecast the ordering of the several hundred different sheet metal blanks.


    The remainder of the product is made to-order by lean manufacturing techniques, such as cutting sheet metal blanks on programmable CNC laser cutters and bending parts on programmable CNC brakes. The flow of information and materials is similar to the general model illustrated in the accompanying article on Mass Customization.

    All parts are made to-order from standard raw materials or cut-to-length as needed. One classic example is the gasket retainer "angles" that are each rolled and cut-to-length from one roll of sheet metal on-demand.

Cut-to-Length operations

    In order to eliminate the need for ordering parts -- and waiting for them -- the following system is used. Linear parts like hinges, hinge pines, and stiffeners, lift angles, and channels are "ordered" (monthly deliveries are arranged) like the standardized sheet metal discussed earlier: as standard parts of the maximum dimension, for instance, the longest hinge. The linear parts are then cut to length as needed. The specific length for a given job is retrieved from on-line monitors. This resulted in consolidating hundreds of discrete components into a few dozen raw material items.

Kanban Resupply

    The remaining parts are resupplied by kanban resupply techniques.3 T he kanban system provides automatic resupply of purchased parts and internally made parts without the delays and costs of purchase orders. This system has worked extremely well in simplifying parts replenishment at Hoffman. Parts are drawn from one of two bins. When the first bin is depleted, it goes back to its source -- either within the plant or outside suppliers. Parts are then drawn from the second bin. The size of the bin has been calculated so that the replenished first bin will return before parts run out in the second bin, even in the highest demand scenario.

Breadtruck Resupply

    Small, inexpensive parts are resupplied by the breadtruck (or free-stock) system, where parts, like fasteners, door hardware, and paint are resupplied by vendors who have been contracted to keep the bins full and bill Hoffman at the end of the month for parts consumed. This system ensures that these part will always be available without any purchasing overhead or delays. A strong relationship is being built with a single supplier for most of the hardware. Limiting the number of suppliers has also helped reduce the amount of overhead required to supply parts to the plant.

The Configurator

    The order process starts with a dialog between the customer and the salesperson using a configurator. Note that this is the only two-way dialog in the system. In the past, Hoffman received a request and then had to check with all the appropriate people in engineering, purchasing, and manufacturing to find out if the job could be done, what it would cost, how long it would take, and what changes to the customerís request would make it feasible.

    The configuration software is really an expert system that contains the knowledge of all the people in manual loops of the old system. It includes allowable options and option combinations, rules, constraints, resources, in addition to costing and delivery algorithms. Thus, the order can be accepted, verified, modified, quoted, accepted, and confirmed almost instantly.

Parametric CAD

    The database then disperses information to a the parametric CAD work station, which has been set up with templates that have floating dimensions for height, width, depth, etc. When the data comes in, it is inserted into the appropriate dimensions and the assembly drawing, and all its parts, automatically stretch reflect an accurate to-scale representation of the product.


    Then the drawing data is converted through CAD/CAM software into machine tool programs for the laser cutter, bending brake, and automatic welder. These programs are transmitted to the appropriate machine tools. Note that specific product "designs" created by parametric CAD and subsequently generated CNC programs are created "on the fly" for single work orders and are not saved. For non-programmable manual operations, instructions need to be conveyed and understood quickly, without the setup to find or understand paperwork.


    Having transformed its engineering and manufacturing processes and dedicated its new manufacturing center in Mt. Sterling, Kentucky, to fast turnaround of its products, Hoffman has been able to delight its customers by consistently delivering custom products within a few weeks of receiving their orders, down from what had sometimes been a ten-week turnaround time. The next challenge will be to take the lessons learned in its Mt. Sterling facility and integrate them into the rest of its operations.


1. David M. Anderson,  "Build-to-Order & Mass Customization, The Ultimate Supply Chain Management and Lean Manufacturing Strategy for Low-Cost On-Demand Production without Forecasts or Inventory" (2008, 512 pages; CIM Press), Chapter 4, "Part Standardization."

2. ibid. Ch. 8, "On-demand Lean Production."

3. ibid. page 232, "Kanban"

     Dr. Anderson is a California-based consultant specializing in training and consulting on build-to-order, mass customization, lean/flow production, design for manufacturability, and cost reduction. He is the author of  "Build-to-Order & Mass Customization, The Ultimate Supply Chain Management and Lean Manufacturing Strategy for Low-Cost On-Demand Production without Forecasts or Inventory" (2008, 512 pages; CIM Press, 1-805-924-0200, and "Design for Manufacturability & Concurrent Engineering; How to Design for Low Cost, Design in High Quality, Design for Lean Manufacture, and Design Quickly for Fast Production" (2008, 448 pages; CIM Press, 1-805-924-0200;  He can be reached at (805) 924-0100 or; web-site:

Call Dr. Anderson at 1-805-924-0100  (Pacific time) to discuss implementing these techniques or e-mail him at with your name, title, company, phone, types of products, and needs/opportunities.

[Mass Customization and Build-to-Order home page]     [Articles Directory]

[BTO & Mass Customization Consulting]  [BTO & Mass Customization Seminars]   [BTO & Mass Customization Books