Biomedical Engineering Reference
In-Depth Information
It was carried out following a request from management, and after the
arrival of a new machine. This machine was entirely automated, was much
faster than earlier machines, and allowed inserting more than five docu-
ments in a single envelope. Unfortunately, the workers proved unable to
bring out the produced mail for several months. Work analysis made it
possible to identify the causes of these difficulties. To set a machine for
folding and inserting the documents in the envelopes, it is necessary to
construct a representation of the kinetics of the paper while the machine
is in operation, in order to see how the paper will be warped. Indeed, it is
impossible to define proper settings for the machine and to avoid paper
jams without being able to do this. However, because the new machine
included a hood, it was impossible to collect this information regarding
paper warps (in spite of the large amount of information that was avail-
able on computer screens). However, work analysis also showed that the
older machines, which had been used before the introduction of the new
automated machine, had been greatly altered by the workers. They had
altered the hoods to add some Plexiglas windows, thus facilitating the
collection of significant information. The alterations that had been made
were therefore analyzed from the point of view of their functions. These
data served as a basis for the construction of functional specifications for
the design of a new automated machine.
In this example, the alterations that workers had made on the machines
were viewed as instrumental hypotheses. They were not implemented
'as is' by the designers. It is the identification of functions and user needs
that is at the source of the transformations carried out by the workers that
is of interest to us. Technical possibilities may allow us to provide novel
solutions, which the workers may not have identified yet. Objectifying the
workers' instrumental hypotheses therefore requires specific work on the
part of the ergonomist, relying on his or her ability to take into account
the level of uncertainty of the design process, and the corresponding work
of designers.
Designing instrumental hypotheses
In this second view, the activity of the designer is (temporarily) in the
forefront. This activity consists in specifying the artifactual side of the
instrumental hypotheses, by supposing, during the design of the artifact,
that its properties may be altered by the workers based on the develop-
ment of their own activity.
Our example will be the LENS software program, used to manage
electronic mail. This piece of software has led to an entire series of stud-
ies. It is a software program that was originally designed as an 'intelligent'
agent. Mackay (1988) has shown that the workers alter the functions
afforded by the software: they wish to be notified of the arrival of new
