Christine V Bullen.

A case study of office workstation use online

. (page 1 of 3)
Online LibraryChristine V BullenA case study of office workstation use → online text (page 1 of 3)
Font size
QR-code for this ebook







Christine V, Bullen
John L. Bennett
Eric D. Carlson

March 1982


Sloan WP No. 1285-82

Center for Information Systems Research

Massachusetts Institute of Technology

Sloan School of Management

77 Massachusetts Avenue

Cambridge, Massachusetts, 02139




Christine V, Bullen
John L. Bennett
Eric D. Carlson

March 1982

CISR WP //84

Sloan WP No. 1285-82

Q C.V. Bullen, J.L. Bennett, E.D. Carlson 1982

Center for Information Systems Research

Sloan School of Management

Massachusetts Institute of Technology


JUL 2 9 1982



Christine V. Bullen

Center for Information Systems Research
Massachusetts Institute of Technology
Cambridge, Massachusetts 02139

John L. Bennett
Eric D. Carlson

IBM Research Laboratory
San Jose, California 95193

ABSTRACT: This paper describes the use of the Office Analysis Methodology to study a
specific office environment in order to determine requirements for an advanced office
workstation. The research site environment was unique in providing an opportunity to
observe a natural growth pattern in the use of advanced technology. Specific workstation
requirements were identified and are being implemented. Interesting observations are
reported in the following areas: categories of secretarial work, use of existing workstations,
influence of a community of users, access to shared services, and impacts on productivity and
organizational behavior.



In the Spring of 1981 the IBM San Jose Research Laboratory (SJRL) was faced with
an operational problem. A variety of typewriters and terminal equipment, installed in offices
throughout the Laboratory, had been acquired over a period of time for use by administrative
and secretarial workers. As part of planning for an expansion of physical facilities, a
committee began investigating how the equipment was actually used in order to make
intelligent recommendations about what new equipment should be provided for the
administrative support staff in the future.

At the same time, a Computer Science group within the Laboratory was developing an
advanced office workstation. It became clear that a study of the work patterns of the
administrative support staff in this particular Laboratory could be helpful in understanding
the general requirements for an advanced workstation.

After reviewing published methodologies and inviting proposals from academic groups
involved with office systems research, the decision was made to work jointly with the MIT
Center for Information Systems Research (CISR) and that funding for the study would be
shared by administrative and research groups at SJRL. CISR was in the process of
conducting research into the nature of office work using the Office Analysis Methodology
(QAM) developed at MIT [ 1 ] . The CISR research is designed to explore office automation
issues in several organizations including samples from manufacturing, high technology, and
service industries.

As Zisman [2] observes, much of the previous work directed toward office automation
has focused on mechanization of the current straight-forward, paper-producing tasks. The

CISR studies, conducted as part of the CISR research on office automation, encompass the
total range of procedures carried out by office workers at all levels - secretarial to top
management. The studies seek to identify, through interviews during site visits, procedures
that have a high impact on the mission of offices within organizations. Once the procedures
are outlined and related to the mission of an organization, then those high impact procedures
and tasks that would benefit most from computer-based technology can be identified.

We present here the results of one part of the CISR study conducted at the SJRL. This
part focused on existing secretarial tasks performed using existing workstations (Section II).
The results (Section III) were used to develop requirements for office workstations for the
SJRL (Section IV). Although the results presented here are from a single case study, they
do indicate the value of using a systematic methodology such as OAM to study office work.
From our study of office environments similar to that found at the SJRL, we beUeve that the
requirements apply generally (Section V) to office workstations in highly automated offices.



Two objectives were chosen for the first phase of the specific study at the SJRL:

A. Survey the tasks currently performed by the secretarial staff in the SJRL.

B. Understand the current use of the existing workstations and the role that they play in
support of these secretarial tasks.

The objectives of the study were focused initially in order to answer the shorter range
operational question of workstation requirements and selection. The interviews were directed
toward understanding the existing procedures and the possible effects arising from the
acquisition and installation of new office workstations.

While we did look for procedures and tasks which could benefit from additional
automated aids, that was not a main focus. The difficult issues of measuring productivity and
predicting the impact of change on the organization were not central in considering what
equipment would be needed to support existing work. However, in the course of the
interviews, factors impacting productivity and organizational behavior were observed.

The Office Analysis Methodology

The Office Analysis Methodology (OAM) was used to guide and structure the study.
OAM focuses on 1) understanding how each office operates within the organization with
respect to the overall organizational mission, and on 2) understanding how that mission is
accomplished. This focus involves conducting a "functional" analysis of the office's
operation, expressed in business terms. The procedures being performed and their purposes
are identified so that analysts, programmers, and office workers can communicate effectively
about requirements.

OAM defines several levels of abstraction as a conceptual framework for gathering
data. The MISSION of an office support group (e.g., the secretarial staff) is described in
terms of purpose and goal (e.g., support the technical staff by preparing documents, handling
phones, and managing office work). A FUNCTION (e.g., document preparation) is the
aggregate of all the procedures that INITIATE, MANAGE, and TERMINATE the use of
office resources to achieve a business goal (e.g., keying, proof reading, printing for review,
and revising text). A RESOURCE is an entity (e.g., a document, a word processor) that is
managed to meet a business goal. A PROCEDURE (e.g., an outline of the sequence for
printing a photo-composed draft) prescribes the tasks needed to complete an activity. A
procedure (or the tasks specified within a procedure) will often involve the manipulation of a

specific OBJECT or set of objects. An OBJECT (e.g., a typed page, an instruction book) is
a tangible entity that is a component of a resource or that provides information about the

0AM offers potential benefits by avoiding the following pitfalls often encountered in
the use of conventional requirements analysis:

- suboptimizing present procedures as a result of a focus on discrete procedures and tasks
taken out of context;

- preserving archaic procedures as a result of a concentration on mechanizing a discrete
process without gaining an understanding of the bigger picture.

In addition OAM provides the opportunity to identify which activities are valuable in
accomplishing the mission, as opposed to identifying only the easily observable, visible,
structured tasks. Through this approach, OAM can help to define productivity and isolate
meaningful measures that apply to semi-structured tasks found in the office, the work as it is
actually carried out [ 3 ] .

The OAM is intended to be quite comprehensive. It can be used for descriptive studies
(how work is done now in a office) or it can be used for prescriptive studies (how new
procedures can be used to better carry out the function of an organization). It includes what
to do in a study, recommends interview procedures, and outlines analyst qualities needed. It
suggests concentration on the usual path through a process followed by analysis of exceptions
and how they are handled.

The best available description of OAM is given by Sirbu et al. [ 1 ]. We can only give
an outline of the concept here. Figure 1 gives an overview comparison and contrast between



look for processes structured
enough to be completely

focus on requirements of
functions within the organization
(not on operational details)

concern with specific
procedures instead of

look for a single system

little attention to
behavioral and managerial
aspects of system design

focus on the need for (low
level) change and the
technology which can be
applied to this end

oriented around functions
and resources which are
then supported in procedures

functions can be supported
by a variety of procedures,
alternative system approaches

concern for decision-making
role of office staff at all

concern with organizational
needs of client group at all

how much secretarial time is
spent typing

how many forms are filled out
per unit of time

how many hours each by how

many people does it take to

complete a procedure;

how often per week is it


how many procedures are in

process at any one time

how many resources are in process
in a unit of time to carry out a
business function

Figure 1. The Office Analysis Methodology permits a focus on functions and procedures
important to the mission of the organization. This contrasts with a conventional approach
focused on the supporting technology.

conventional requirements analysis and OAM. Figure 1 is not meant to be a comprehensive
comparison; it only highlights some key differences.

We began our preparation for the study at San Jose by adapting an interview outline
previously developed for use as part of the CISR study at other sites. Figure 2 is a summary
of the interview outline.

The Case Study Site

The IBM San Jose Research Laboratory (SJRL) is organized into four major research
areas, called Functions,^ each of which is headed by a Functional Manager who reports to
the Director of the Laboratory: Computer Science, Physical Science, Storage Systems, and
Applied Science. All centralized administrative tasks are combined at a Functional level in
Administrative and Technical Services (A&TS). The total population - research and support
personnel - is about 600 people.

Within each research Function there are two to three departments ranging in size from
23 to 45 people. The people on the research staff, called Research Staff Members (RSMs),
are grouped by research specialty within the departments and are assigned to specific
projects. While the research staff is relatively stable, shifting between departments and
projects is quite common. In addition, a number of visiting scientists, postdoctoral fellows,

^The Use of the term FUNCTION is not identical to the use of the term in OAM. However,
both uses relate to a focus on activities needed to achieve mission results.



Statement/organization chart


A. People (who, how many, management levels)

B. Other resources


A. Phases (initiating, managing, terminating)

B. Inputs/Outputs

C. Sources, destinations, links

D. Exceptions (a checklist of exception causes was provided)

E. Objects (a checklist of si mple objects was provided)

F. Databases (a checklist of sample manual and electronic

data bases was provided)

G. Quantitative measures

1. Number of objects in process at any time

2. Time to accomplish a task

3. Time to accomplish a procedure (set of tasks)

4. Frequency of repetition

5. Number of objects processed per unit time

6. Timing constraints on completion of a task

7. Frequency of exceptions

8. Number of people involved in each step of a procedure

9. Size of databases

H. Office Layout/Environment

1. Equipment (what, what used for, likes and dislikes)

2. Comfort, style

3. Training

4. Special needs

Figure 2. The outline structured the Phase 1 interviews such that the information gathered
on clerical procedures could be understood in the context of the office and organization

and summer interns report at functional, departmental, or project levels while temporarily at
the SJRL.

In this phase of the study, we confined interviews to the secretarial staff working at the
Function and Departmental levels. We also included those secretaries working in the
Administrative Processing Center giving support to A&TS professionals.

The SJRL provided an intriguing research site. While it is comparable to our other
research sites in its basic organizational design and its administrative functions, it is unique in
the following ways. First, the SJRL is staffed by technology-oriented, highly skilled
professionals who create an environment which is receptive to introduction of new
technology. This would be expected in any group working on advanced technology products.
However, the innovative attitudes of the RSMs encourage experimentation with new
technology even in their routine office activities - drafting papers, preparing presentations,
and sending messages to colleagues. In addition, sophisticated technology is available for use
by administrative people in a setting without mandate or formal pressure to employ it. This,
combined with the supportive access to information from RSMs, results in an unusual
situation for studying natural growth patterns in the use of advanced technology. Thus, we
could observe actual patterns of use in a technologically sophisticated envirormient as
contrasted with the speculation common with many writers on the "office of the future."

As outlined above, we used the OAM framework to construct a one hour interview.
The categories of those interviewed and the range of people nominally served by each
secretary in the category are as follows.

Number of people

Served by each



Number of




itional Manager


irtment Manager






3 to 7

23 to 43

5 to 45

Each secretary was interviewed at the place of work where sample objects
(computer-readable and paper copy) could be displayed as needed to serve as an illustration.
All secretaries interviewed used a desk-top display terminal capable of showing
simultaneously on the screen 24 lines of 80 uppercase and lower case characters. Some had
a terminal allowing display of 43 hnes at a time. Each terminal had an attached (but
movable) keyboard. In addition, each secretary or administrative support person had a
communicating tjrpewriter terminal for printing output on letterhead paper. This terminal
was also used occasionally as a stand-alone typewriter, and it had magnetic card storage.

The display terminals, which had no stand-alone data entry capability, were attached to
a large-scale host computer operating the VM/CMS system. Also attached to the computer,
directly and through a network, were many high speed and/or high quality devices used to
produce printed output on a variety of paper and pre-printed forms. The network is an IBM

''The selection of those interviewed and the range estimate for the number of people served
were made from an inspection of the organization chart.


corporate network linking computers in most IBM Laboratories worldwide. The secretaries
in the sample were using a variety of software available on the system, including a full-screen
editor, a document formatting and printing facility, a message system, and a number of
locally-developed macro programs.


Types of Tasks

In our study we focused on resources and objects that resulted in paper copy or that
went through a keyboard data-entry phase. We did not address phone handling as a task
(except to note approximate percentages of time spent), although we did consider typed lists
as support for making phone calls.

The secretarial work at SJRL can be divided into the two categories shown in Figure 3.
The "others" initiating secretarial work are professionals, managers, and visitors serving on
the staff. The category of "work initiated by others" is the one that generally comes to mind
when people describe what the "typical" (actually stereotypical) secretary does. The work is
text-oriented - that is, the initiator provides text (handwritten, dictated, rough-typed), and
the secretary's role is to provide text output in typewritten or printed form. Completion
requires little contact with other resources (documents or people). The tasks require a fixed
format which is made standard through policy, tradition, or equipment constraints. The
outputs are typically a file specifically designed to be revisable (because the final task result
is subject to initiator negotiation) and text-on-paper for initiator review. Examples are
notices, letters, and memos (relatively short) and activity reports (relatively long). In the
second category.



1 . Work initiated b;;^ "others" secretary is

a. told explicitly what to do

(given raw text and a sample letter specifying the format)

b. given some discretion

(foil format)

2. Work initiated b^ the secretaries in response to:

a. being told in a general sense to achieve a result

(produce an equipment inventory list)

b. being told they are responsible for a result

(making labels used to forward mail to former visitors)

c. observing a need and taking responsibility for meeting it

(monitoring department expenditures)

Figure 3. Use of the OAM led to insight on process-initiation as a key discriminator of the
type of work done by a secretary.

the levels of self-initiated work reflect increased taking of responsibility and creativity.
Although this category involved keyboard data entry, the data entry is not an end in itself.
Completion generally requires contact with others, and the secretary has flexibility in
selecting the format in which the results are presented. The outputs are typically files used
by the secretary in carrying out office procedures. Examples are mail logs and reminder files.

There seem to be a series of prerequisites for appearance of secretarial work in the
"self-initiated" category. First, powerful tools (or a light work load) must make it possible
to get routine work out of the way. Then the secretary must have a willingness to explore


the use of tools in imaginative ways. Third, the professionals served must acknowledge the
value of the resulting innovation.

Hiltz and Turoff [4] describe the category of "work initiated by others" well when
they observe that secretaries "act as intermediaries between the originators and the recipients
of text." They comment on the fact that word processors are typically aimed for one
specialized aspect of what the secretary actually does.

There are other non-text kinds of work in the "initiated by others" category (e.g.,
placing telephone calls). The extent to which a secretary can influence the process used to
complete such tasks depends on the precision with which the "order" is given. The initiator
may fully describe the task (e.g., specify the exact format) or may leave that to the discretion
of the secretary. An instance where the secretary adds value to the output is in the design of
overhead projector foils. In this category, there are two breakdowns: those tasks where the
value-added is in form only, and those where both form and content are influenced by the
secretary. Figure 4 summarizes some examples in each Ust.

Tasks initiated by secretaries, in order to better accompUsh a job, involve a flexible
format. Examples include equipment inventory, personal calendars, and budgets. The
self-initiated work described in Figure 5 includes use of tools requiring the secretary to do
considerable "thinking"; that is, the secretary must add value in order to achieve the result.
Again, many more "non-keyboard" tasks could be found in this category when observing a
secretary at work (e.g., telephone calls, setting up meetings, planning office moves, furniture
and equipment acquisition). Figure 5 lists some examples of self -initiated tasks requiring
keyboard data entry.




typing a letter

technical typing

(formulas, equations,
special characters)

foil layout

activity reports

progress reports
performance plans
research orders

memos announcing a meeting

applicant handling

speaker announcements

shaping notes into
sentences and paragraphs


space planning, moves,
and telephone assignments

new staff, visiting pro-
fessionals, summer interns;
orientation and records

equipment inventory

Figure 4. Tasks initiated by others but which give evidence of secretarial "value added" and,
in some cases, creativity.


mail log

distribution lists for reports and memos

equipment inventory

employee home addresses and phone numbers

mail forwarding lists and labels

reminder file based on date

financial monitoring to track budget expenses

Figure 5. Tasks initiated by secretaries in order to support their own work, to accommodate
their personal working style, or in response to an enviroiunental need they have observed.


A major result of our interviews is the observation of how time is allocated between
categories of work. The secretaries to Function Managers spend less than 50%*^ of their
time doing structured text entry initiated by others. Secretaries to Department Managers
spend from 50% to 90% of their time doing such work, depending on the style of the
Department and the style of the individual secretary. The remaining secretaries spend 75%
to 95% of their time on this category of tasks.

It is clear that much of the secretarial workload consists of tasks initiated by the
secretary. The conventional stereotype is that secretaries only carry out highly structured
tasks at the direction of the persons supported. Contrary to the stereotype that "secretaries
do not make decisions of any significance on how to spend their time," we found that
secretaries to Function Managers typically spend more that half their time working on
self-initiated tasks, and Department secretaries spend anywhere from 10% to 50% of their
time in such activities.

The Nature of the Documents

In the previous discussion of types of tasks, the variety of documents at the SJRL was
also illustrated. Figures 3, 4 and 5 all list specific examples of the documents identified in
the course of the study. Because document preparation, storage, retrieval, and printing are
major office workstation tasks, we used the OAM "objects" concept to investigate how
documents enter into the SJRL secretarial work flow. We categorized documents on the

^he time estimates were collected in the interviews and reflect the judgment of those
interviewed. The figures were not independently validated. However, the results are
consistent within the hierarchy of secretaries, suggesting that these approximations are
reasonable. The interview results were supplemented by some direct observation.


basis of frequency as seen by the secretary. Figure 6 shows this classification with another
sample of documents, preprinted forms, listed in each frequency category. The study
identified over 50 different (in terms of format) documents, about equally divided among the
three categories. Document preparation can be either self-initiated or initiated by others, but
it is low on the scale of value-added. However, document preparation may be a task
associated with a much more significant procedure (e.g., planning for and ordering new office
equipment) that may be important to the mission of an office.

The Use of Existing Workstations

Each secretary had access to the display terminal that was connected to the host
computer and to the communicating typewriter terminal with printer and magnetic card
storage. We were interested in the secretarial preferences for choice of use and the criteria
entering into the decision. We expected that these preferences would illustrate characteristics

1 3

Online LibraryChristine V BullenA case study of office workstation use → online text (page 1 of 3)