Mechanical Technology January 2016

36 / 44

34 Mechanical Technology — January 2016
⎪ Structural engineering materials, metals and non-metals ⎪
The Chrysler plant in Kokomo,
Indiana produces gearboxes.
This includes design, casting,
machining and assembly. Chrys-
ler’s philosophy of placing its best and
brightest on the factory floor is probably
not unique and similar practices support
Japanese manufacture, but in South
Africa, the factory floor is not as highly
regarded.
The rationale is that, as monies are
made or lost on the factory floor based
on the performance of the end product in
the market, looking for nascent problems
before they emerge and for opportuni-
ties for improvement make commercial
sense. The administrative offices in these
facilities are sparsely occupied, with all
the necessary performance communica-
tion reduced to dashboards.
Locally we seem to prefer a top-down
approach, with substantial administra-
Materials engineering in practice:
the product-centred approach
In this month’s column from Wits’ School of Chemical and Metallurgical
Engineering Tony Paterson discusses the advantages of moving from a
sequential approach to fabrication to a product-centred approach, based on
reciprocal interdependence between the parties involved at every stage, from
design, though material processing and manufacture.
Figure 1: Sequential interdependence between independent work centres. Coordination is achieved through planning and control via contracts.
Figure 2:
The product-centred
approach and reciprocal
interdependence enables
each party to be dependent on
each other. Coordination is achieved
through mutual cooperation and
adjustment based on end product goals.
tive offices and/or contract-based rela-
tionships. This represents a sequential
interdependence between discrete work
centres, where coordination is achieved
through planning and control, often
through rigid procedures. This, illustrated
in Figure 1, results in contractual rela-
tionships with independent work centres
linked via contracts. It works reason-
ably well in an industrial production
environment.
Sequential interdependence, however,
rarely results in effective communication.
This is required for effective performance
in project engineering passing through
workshops and jobbing shops where
products vary. Figure 2 shows an alter-
native product-centred approach. The
end product is the focus of success for
all technical decisions. The correspond-
ing communication structure is shown
interacting with the product and with
all other work centres. The key issue is
the need to achieve mutual cooperation
and adjustment between all the centres
of expertise.
Why do we not follow this route?
Whilst engineering is taught as a science,
it is in fact an applied science. The sim-
plified models taught may be simple or
complex to analyse but usually fall into
the known-known box of Figure 3. These
models form a valuable reference base
from which to work. However, applied
science represents the art of engineering.
The key skills required are judgement
and compromise. Judgement considers
both theory and applied knowledge,
while compromise is necessary where
required outcomes clash. Judgement is
almost always required when unforeseen
operational circumstances arise.
In the context of materials, for exam-
ple, operational circumstances define the
load and load effects to which a structure
will be exposed during its working life. In
essence, anything that results in a stress
in the material may be regarded as the
material response to a load. However,
in practice it would appear that, par-
ticularly with the advent of systems,
computer assists and specifications, the
applied science that is engineering is
being regarded as a pure science. For
instance the Chemical Manufacturers’
Association defines mechanical integrity
as “the establishment and implementa-
tion of written procedures to maintain
the on-going integrity of the process
equipment.”
technical-knowledge]. This supports
the sequential interdependence concept
shown in Figure 1.
Whilst system thinking, computer
models and appropriate input or output
specifications are valuable tools which
should be embraced, they do not replace
the need for expert input, this coming
from a range of disciplines as we seek to
develop lighter structures. Metaphorically
tossing a problem over a contractual wall
may only serve to shift blame rather than
to gain from the opportunities of col-
laboration using expertise from various
sources. What makes engineering con-
tinually interesting is that it is an applied