Civil Engineering Reference
In-Depth Information
1. There is no relative motion between the two parts
2. The materials of which the two mating parts are composed do not have to be different
For each part the assembly time is measured. Boothroyd then makes the assumption that
an ideal time for a part is 3 seconds. This is reasonable for a part that is easy to handle
and insert. A measure of the manual assembly design efficiency
(E
m
)
is then obtained
using the equation:
E
m
=
3
N
m
/T
m
where
N
m
is the minimum number of parts, and
T
m
is the total assembly time. If
E
m
<
1
then the design is inefficient, and if
E
m
>
1 the design is efficient. The value of
E
m
is,
however, not always conclusive. Complex electromechanical products that require
extensive wiring tend to have low design efficiencies, even when well designed. On the
other hand, simple products with few parts can have high design efficiency. In their
handbook, Boothroyd and Dewhurst (1987) provide many examples of successful
redesign where productivity gains of 200-300% were obtained.
17.5
MTM ANALYSIS OF AN ASSEMBLY
Boothroyd's technique is useful for redesigning existing products, but it cannot be used in
the design of new products at the conceptual stages of design. Predetermined time-and-
motion studies (PTMS) can be used for this purpose. As a basis for our analysis, we use
motion time measurement (MTM) (e.g. Konz, 1990). In MTM, an assembly is broken
down into several constituent tasks, including reach, grasp (pick up and select), move,
position part, and insert. MTM specifies the amount of time it takes for a trained worker
to do each of these elemental tasks. However, the assembly time depends very much on
how the product is designed. Table 17.2 illustrates time savings for a best design case as
compared with less efficient design. For example, reaching to a fixed location is the best
case and takes about 30% less time than reaching to a variable location or to small and
jumbled parts. Grasping of easily picked up objects is 75% faster than for objects on a
flat surface. Hence the design engineer should design parts that are easily reached and
easily grasped. The parts should be presented at a fixed location. This can be
accomplished by using part feeders (Figure 17.11). Much research has been performed to
develop part feeders for robots (Boothroyd, 1982). These can also be used for manual
assembly. A cost-benefit calculation can easily determine whether parts feeders for a
manual assembly are cost-efficient. Simply calculate the time savings for assembly and
compare to the cost for parts feeders.
Following the “pick-up” the part has to be transported and positioned for the final
insertion step. Table 17.2 illustrates that moving a part against a stop (case A) requires
about 15% less time than when a part is moved to a location without a stop (case B). In
the latter case the absence of tactile feedback requires greater manual control. Ironically,
most products are assembled as in case B. One objective of good design must therefore
be to incorporate stops which provide tactile feedback (Furtado, 1990).
The time to position parts depends on whether the part is symmetrical (code S) or non-
symmetrical (code NS). In the latter case the operator must turn the part, which takes
30% longer. To insert a 4-in part takes twice as long as to insert a part with no depth. If
