22

LOGISTICS INFORMATION MANAGEMENT 5,1

roject scheduling has advanced, particularly

in a job-shop or order-driven environment. Data processing provides the most
dramatic evolution in practice, with desktop
potential now changing the role of managers.

New
Developments
in Project
Scheduling
Fritz P. Egger and Brian H. Kleiner

Logistics Information Management, Vol. 5 No. 1, 1992, pp. 22-24

MCB University Press, 0957-6053

Research into modelling and management techniques

continues. The arrival of artificial intelligence and the
development of the project management profession are
further advances, both now and in the future, that will
refine project scheduling[l,2,3,4]. The data processing
revolution has been the primary source of improvement;
however, increased management skill and behavioural
understanding will also result in the realization of gains.
Scope
The general scope of project scheduling is quite wide. This
article examines the topic in the context of a job-shop
environment in which an engineered product is
manufactured. This type of operation usually contrasts
sharply with batch and process manufacturing.
Scheduling is an imperative part of the project management
process. Good planning is essential to any successful
execution of a schedule. The items to be scheduled are
factors of production such as human resources, machinery,
capital, and the individual tasks and activities involved with
the former factors. Careful attention must be given to the

breakdown of the tasks commonly known as the work

breakdown structure.
What makes a successful project? Traditionally, a project
and its manager have been considered successful when
the project is completed on time and within (or under)
budget, without sacrificing quality. The manager must
decide whether the project is budget driven or time driven.
Creating a comprehensive plan (schedule) is critical to the
success of any project, large or small, simple or complex.
According to Peter Drucker, in The Effective Executive[5],
time is a unique resource. Of the other major resources,
money is actually quite plentiful. People, the third limiting
resource, one can hire though one can rarely hire enough
good people. But one cannot rent, hire, buy or otherwise
obtain more time.

Medieval Remains in Project Management

Many antiquated ideas are still found today in project
management and are even integrated into computer
models. According to Lichtehberg[6], the two most
commonly remaining factors are:
(1) The prestige of the exact figure.
(2) The dominance of objective matters.
Project management contains many undefined areas which
are often erroneously converted into hard figures. This
has a negative impact on the schedule. Object matters
still tend to be dominant in people's minds. The insistence
on exactly repeatable experiments has carried over from
the natural sciences but yields excessive detailing and
schedules that cannot be realistically achieved.
Computer-based Developments
Computer-based assistance for the scheduler has come
of age. Software programs for personal computers, as well
as for mid-size machines, abound. Two such programs
are explored in this article.
The computer's decision-making process is based on the
work breakdown structure. This technique divides blocks
of works down to the point of manageability, which allows
them to be identified and scheduled as individual tasks.
Ronan[7] defines the ultimate breakdown as "the lowest
level of controlled effort". The resulting tasks are often
graphically represented on GANTT charts. The beauty
of these charts lies in their simplicity, although greater
detail may be added.
Decision trees, networks, critical path method (CPM) and
Program Review Evaluation Technique (PERT) have been
in use for decades. PERT shows the relationship among
tasks, while CPM displays the path showing the critical
items that the project's schedule hinges on assuming no
slack. Today's modern computer programs focus on
resource levelling and management. The work breakdown

NEW DEVELOPMENTS IN PROJECT SCHEDULING

structure can be displayed graphically through GANTT

charts, histograms, pie charts and many other
representations.
Currently, one of the most popular personal computerbased software packages is "Timeline", manufactured by
Symantec in Northern California, who claims to have over
60,000 users. Resources can be levelled to various degrees
and tasks can be prioritized. The entire schedule, which
is resource driven, can be calculated or recalculated within
a few minutes. Cost tracking is also a standard feature.
A calendar that automatically schedules holidays,
weekends and overtime is maintained on a real-time basis.
Tasks can be set up as dependencies of other tasks.
Printed reports are available in the forms of GANTT
charts, histograms, trees and networks, and the entire
program can be run on a personal computer with 512K
RAM memory!

The optimal approach is to

address bottlenecks in a
finite matter
Today's popular mid-size computer benefits from a
multitude of manufacturing packages available in the
market. IBM's AS/400 and Hewlett-Packard's HP1000 to
HP3000 Series offer solutions for prices deemed
unthinkable just a few short years ago. One of the best
packages for the AS/400 and HP3000 for a job-shop
environment, is "Jobscope", which was developed by
Jobscope Corporation in Greenville, USA. Jobscope's
modules cover the gamut of a company's activity, from
order entry all the way to shipment. Included are
comprehensive systems for financial and cost accounting,
as well as for advanced project and shop scheduling. Both
finite and infinite scheduling are allowed, as well as a
combination of the two. The optimal approach is to address
bottlenecks in a finite matter while scheduling other areas
infinitely. Finitely loading all resources would lead to
underutilization of bottlenecks during high demand
periods. The combination approach results in logically
developed, achievable schedules. A simulation mode is
also available, aiding in the identification of bottlenecks
and for conducting "what if" analysis. Forward scheduling
calculates start and stop dates, using a beginning date and
working forward into the future until a completion date
is defined. Backwards scheduling, on the other hand,
calculates start and stop dates from a predefined delivery
date, working backwards to a beginning date. Forward
scheduling tends to produce more realistic results, but
the backwards approach may be necessary for high
pressure projects.

23

Most sophisticated programs incorporate all of the above

features and capabilities. Prior to the availability of today's
electronic aids, scheduling was a tedious, time-consuming
and manual task. We are now able to run an unlimited
number of possible scenarios allowing the best possible
approach to surface.
New Management Techniques
The conceptual side of project scheduling has also given
rise to new developments. Several new models have been
devised:

UC Irvine's Dr Kut So[8] conceived "The Greedy

Heuristic", a technique which selects the most
rewarding unassigned job and assigns it to a
machine with available capacity until no more jobs
can be assigned to any machine. This approach can
be written into software to achieve optimal results.

Dr R. Ramasesh of Texas Christian University

conducted a survey of simulation research on
dynamic job-shop scheduling[9]. He determined
that the Shortest Processing Time (SPT) expediting
method is the best rule for minimizing flow time,
waiting time and number of jobs in work in process.

A data-driven modelling study was conducted by

Lori Franz of the University of MissouriColumbia[10]. She points out that it is vitally
important to relinquish control over departmental
models to the users immediately after implementation. The models must be understandable to
the users to ensure their continued use.

Northern Telecom Inc. has developed a "Gate

Review" technique[11]. The focus is on a review
process which evaluates the effectiveness of
communication between the project team and other
groups, thus aiding in preserving the integrity of
the schedule. A "gate" is a point of transfer
between two phases of the project and serves to
eliminate potential communications problems before
they occur. Gate reviews are most useful for
projects that appear to be heading off course.

The Coming Impact off Artificial Intelligence

Through the use of artificial intelligence, revolutionary
improvements have been made in the fields of medicine
and geological exploration. Many other disciplines,
including project scheduling, should reap similar benefits
in the coming years. Artificial intelligence expert systems
would add analysis and predictive capabilities to scheduling
systems already in place.
The system would be able to monitor deviations from the
planned schedule, signal potential delays, point out the

24

LOGISTICS INFORMATION MANAGEMENT 5,1

needs for human intervention and analysis and show where

to allocate additional resources. This would lead to
improved efficiency and resource use.

Developing the Project Management

Profession
Projects are becoming more complex and are coming
under increased scrutiny. One of the new areas of focus
is the project manager himself. By improving his or her
skills, scheduling, along with the other management
functions, can be improved. This is especially true in the
international context, where a great variety of behavioural
patterns must be dealt with effectively.
Internet International[12] is a 25-year-old organization that
was created to facilitate association and interaction
between project management experts, interested parties
and trainees. World congresses are held every three to
four years. A different topic highlights each gathering. In
1985, 14 countries and more than 4,000 members were
represented in the organization.

Summary
The most dramatic developments in scheduling are the
results of the data processing revolution. Activities and
analysis that once required huge computers can now be
handled from personal computers that fit on a desk top.
Company-wide software packages that integrate all details
such as vendor lead times, shop backlog, critical paths,
bottlenecks and other important factors can now be run
on reasonably priced hardware.
The impact of artificial intelligence systems will lead to
continued increases in efficiency and predictability.
Similarly, the development of project management as a
profession will lead to increasingly skilled schedulers and
analysts.

References

1. Avots, I., "The Coming Impact of Articifial Intelligence

on Project Management", in Grool, M.C. (Ed.), Project
Management in Progress: Tools and Strategies for the 90s,
Elsevier Science Publishers, Amsterdam, 1986, pp.
307-12.
2. Bernard, P., "Inaccurate Schedules? Check your CRP
Logic", P&IM Review with APICS News, Vol. 10 No. 1,
1990, pp. 37-40.
3. Owens, D. and Martin, M.D., "Project Management and
Behavioural Research in an International Context", in
Grool, M.C. (Ed.), Project Management in Progress: Tools
and Strategies for the 90s, Elsevier Science Publishers,
Amsterdam, 1986, pp. 141-9.
4. Woodward, J.F., "Development of Project Managers from
the Year 2000", in Grool, M.C. (Ed.), Project Management
in Progress: Tools and Strategies for the 90s, Elsevier
Science Publishers, Amsterdam, 1986, pp. 67-72.
5. Drucker, P., The Effective Executive, Harper & Row, New
York, NY, 1967.
6. Lichtenberg, S., "Medieval Remains in Modern Project
Management And Some Successor Principles for the
Nineties", in Grool, M.C. (Ed.), Project Management in
Progress: Tools and Strategies for the 90s, Elsevier Science
Publishers, Amsterdam, pp. 57-65.
7. Ronan, D.D., Managing Projects: A Systems Approach,
Elsevier Science Publishing Co. Inc., New York, NY, 1986.
8. So, K.C., "Some Heuristics for Scheduling Jobs on
Parallel Machines with Setups", Management Science,
Vol. 36 No. 4, 1990, pp. 467-75.
9. Ramasesh, R., "Dynamic Job Shop Scheduling: A Survey
of Simulation Research", Omega InternationalJournalof
Management Science, Vol. 18 No. 1, 1989, pp. 43-55.
10. Franz, L., "Data Driven Modelling: An Application in
Scheduling", Decision Sciences, Vol. 20 No. 2, 1989, pp.
359-76.
11. Garrette, L., "Rescuing Development Projects with Gate
Reviews", ASM/Journal of Systems Management, Vol. 41
No. 4, 1990. pp. 8-10.
12. Gutsch, R., "Twenty Years of Internet International",
in Grool, M.C. (Ed.), Project Management in Progress,
Tools and Strategies for the 90s, Elsevier Science
Publishers, Amsterdam, 1986, pp. 328-32.

Fritz P. Egger and Brian H. Kleiner are at the School of Business Administration and Economics at California State
University, Fullerton, USA.