There are numerous academic articles and industrial news sources that present successful redesign case studies, case studies where real-world products have been remodeled to significantly reduce parts count and assembly time. Below, a few are presented together with a case study compilation carried out by Booth-royd Dewhurst Inc. on their own customer base. This to give us an idea of what benefits have been achieved using efficient industrial design methods.
We should do well to remember, however, that any articles written about DFA case studies will be predominantly positive simply because of human psychology – it is unlikely that anyone will publish an article about unsuccessful redesign, or
at least about very marginal results from the redesign venture. Small results are most often deemed lacking in news-value.
Case studies
We already remember the two studies presented in connection with the DFA techniques earlier on – Stone et al. (2004) described the redesign of a heavy duty industrial stapler, and the University of Hull present an improved windshield wip-er motor. In the case of the staplwip-er, the authors suggest that it is possible to reduce the part count from 29 to 11 (a reduction of 62%). In terms of assembly time, the old model consumed 204,18 seconds while the new model would use only 88,08 seconds (a reduction of 57%). The windshield wiper motor achieved a parts re-duction from 29 to 6 (a rere-duction of 79%).
Ardayfio, Paganini, Swanson & Wioskowski (1998); Matterazzo & Ardayfio (1992), Ardayfio & Opra (1992), Ardayfio, Dembsey, Kreucher & Schmitt (1998); Sarmento, Marana, Ferreira-Batalha & Stoeterau (2011) da Conceicao &
da Silva (2010) present successful case studies of Automotive Fuel Intake Covers, Automotive Electrical and Electronic Systems, Front Suspensions, Convertible-Top Vehicles, Brake and Clutch Pedal System at Ford Motor Company, GM, Chrysler and Delphi Automotive Systems.
Boothroyd et al. (2002), though hardly impartial in the matter, produce an impres-sive list of successful projects, where the DFA techniques have produces signifi-cant positive results: Donnell Douglas used DFMA methodology to redesign the F/A Hornet jet fighter, resulting in a bigger, lighter product with 42% fewer parts.
Dell Computer Corporation used DFA tools to design and manufacture a new desktop computer chassis, with savings reported as a 32% reduction in assembly time, a 44% reduction in service time, a part count reduction of 50%, and a direct labor cost reduction of 80%. Motorola, on the redesign of their walkie-talkie ve-hicular adapters, realized assembly time reductions of 87%, from 2742 seconds (45,7 minutes) initially to only 354 seconds (5,9 minutes) when finished. Ciba Corning Diagnostics Corp, a medical company building blood-gas analyzers, re-duced their overall number of parts by 48% and cost by 22 %. Magna Interior Systems Seating Group, a company making seats for the car industry pulled down total parts count form 105 to 19 and assembly time form 1445 seconds to 258.
Whirlpool used the method to make their new microwave oven faster to assemble (by 26%) despite adding a new forced convection technology to it, in addition to the old functionality. At Ingersoll-Rand Corporation's portable compressor divi-sion, the DFMA software was used to redesign an oil cooler and a radiator sembly to achieve a parts count reduction from 80 parts to 29 (64%) and an as-sembly time reduction from 18,5 minutes to 6,5 minutes (65%). In a later attempt,
the same company redesigned a control and instrument panel assembly to achieve a parts count reduction from 36 to 24 (33%) and an assembly time reduction from 8.5 minutes to 6.1 minutes (28%).
Port (1989) relates two successful case studies: IBM's redesign of their ProPrinter and NCR's cash register terminal. The IBM printer (automated assembly) was redesigned to result in a parts count reduction from 152 to 91 (40%) and an as-sembly time reduction from 30 minutes to three minutes (90%). Furthermore, the printer could now be manually assembled, because of its more simplistic design.
The cash registered achieved a parts count reduction of 80% and an assembly time reduction of 75% through the DFA analysis. As an added bonus of the re-duced part count, the number of suppliers needed went down by 65%.
Stone et al. (2004) presents the redesign of an ordinary kitchen wok. As for the wok, parts count dropped from 33 to 13 (a reduction of 61%) and assembly time from 233.48 seconds to 91 seconds (a reduction of 61%). Causey (1999), using the DFMA methodology, suggests redesigns to a normal flashlight achieving a part count reduction from 14 to 7 (50%). Stevens & Eijsink (1994) describe the redesign of a wall-mounted emergency fire hose reel. According to their calcula-tions, their redesign reduces parts count from 59 to 27 (a reduction of 54%) and manual assembly time from 525 seconds to 264 seconds (a reduction of 50%).
There is no shortage of impressive cost reductions to be seen in literature – the presented studies are just a sample from the top of the pile and there exists a mul-titude of similar case studies in design literature and journals. The following ex-ample is especially interesting in that it describes the usage of Design for Assem-bly methods already during the creation stage of a new product – a new product only exists in potentia, but cost estimates and assembly simulation can still be done during the development phase to see whether the new design shows promise or not.
Hydrogen fuel cells: using DFA before production
In an article for Design Engineering news (2007) we can read about a project un-dertaken by the technical consulting firm Directed Technologies Inc. on behalf the U.S. Department of Energy, to create a hydrogen fuel cell. The fuel cell should be a tangible alternative energy solution, with set technical and financial goals for each step of the project. The final target (for 2015) is set at 15 dollars per kilowatt, an ambitious goal, considering that the 2006 cost target was set at 70 dollars per kilowatt.
Importantly, the goal of the project is to consider all possible solutions (systemat-ic, material, operational, etc.) that might lead to cost reduction compared to the current products. DFMA software is used to rapidly calculate estimates of manu-facturing and assembly costs with only light initial information.
The hydrogen fuel cell stacks consist of several hundred plates with differing electrical charge that act as anode and cathode, as well as a similar amount of semi-permeable membranes that guide the electrons and protons in a desired route. The design of the plate for optimal processing surface to minimum manu-facturing cost is therefore of great interest to the designing engineers.
After several different iterations, considering the effects of process, plate design, plate material and production batch sizes, it was concluded that stamped steel plates would function better than injection moulded composite plates, and that pieces stamped from a roll of steel with the exact width of the plate would pro-duce zero waste at an 80 plate per minute rate. By changing their designs in this manner, the designers have been able to bring the projected cost of the fuel cells down to $25/kW.
Survey work
There are very few surveys made on the subject of how often redesign projects have been successful, as opposed to only achieving moderate or marginal im-provements. The closest thing we can come to an industry-wide benchmark of the situation in companies where Design for Assembly is used is another study by Boothroyd Dewhurst, Inc. (2004). Again, we must be aware that private company is not the best source of objective data, but since similar studies are almost non-existent, we have to consider it the best source available.
The Boothroyd Dewhurst study compiled the results of 117 product-design case studies by 56 manufacturers, conducted over the last 15 years. The their numbers showed, for instance, that using the DFMA techniques resulted in an average parts count reduction of 54% in 100 of the cases studies. Similarly, an average assembly time reduction of 60% was calculated on the basis of 65 case studies.
The full survey is presented in Table 3. The figures indicate that there is a consid-erable amount of value in the judicious application of Design for Assembly tech-niques.
Table 3. Case study compilation results.
Category Number of
cases
Average reduction (%)
Part count 100 54
Assembly time 65 60
Product cost 31 50
Assembly cost 20 45
Assembly operations 23 53
Separate fasteners 20 60
Weight 11 22
Labor costs 8 42
Manufacturing cycle 7 63
Part costs 8 52
Unique parts 8 45
Assembly tools 6 73
Material cost 4 32
Number of suppliers 4 51
Manufacturing steps 3 45
Assembly defects 3 68