Preparedness and Mobilisation - The Role of Systems Dynamics
in Managements Approach to Resource Allocation
David Paterson
School of Civil Engineering, Australian Defence Force Academy, Canberra ACT 2600,
Tel:(06)2688359, Fax:(06)2688337, e-mail: dj-paterson@ adfa.oz.au
Aim
In a complex, vertically integrated organisation it is often difficult for individual parts to
recognise and maintain focus on a wider common goal. The aim of this study is to describe how a
highly complex problem in a system which crosses several organisational domains has been
addressed through system dynamics modelling and a gaming interface. The defence context
addresses a shortfall in analysis of military preparedness where previous modelling, which
concentrates on the period after identification of a specific threat, is not sufficient for planning in
the Australian strategic environment.
Defence Mission
“The Mission of the [Australian] Department of Defence is to promote the security of
Australia and to protect its people and interests. It does this by maintaining the military capability
required to implement the strategic guidance received from Government. This capability is
“ achieved through a combination of force structure and preparedness of that structure for
operations.
Defence, as a major consumer of public resources, is under continual scrutiny for
budgetary savings. It is critical that defence planners can demonstrate not only that the armed
forces are able to meet the strategic requirements of government, but that this is being done in a
cost effective manner. Conversely, they need to be able to demonstrate the implications of
proposed budget cuts.
Strategic Parameters
In essence Australia’s defence planning since 1976 has assumed no strategic threat from
any identifiable source within a (rolling) 10 year time frame. Around this policy a number of
769
scenarios have been developed as the basis for contingency planning. These scenarios, which
include support to the UN, have varying degrees of intensity, duration, and lead time.
Management Environment
The functioning of any Defence unit is influenced by complex interactions between;
resource decisions, personnel management, logistics management, and training doctrine etc. Each
of these areas, however is likely to be subject to separate and discrete policy development and
resourcing processes. This problem is exacerbated where peacetime resourcing processes for many
elements are not applicable during preparation for or after deployment.
Finally, many decisions take years to have an impact. Equipment acquisition may take up
to 10 years for an off the shelf purchase (much more if the asset is locally developed). Decisions to
change the mix of officer commissioning source may take 4 years to have an effect (the minimum
time to attend and complete study at the Australian Defence Force Academy).
System Dynamics Modelling
The ability of System Dynamics to model feedback with delay would appear to make it the
ideal tool to examine this problem. Unfortunately there are a number of issues that affect both the
complexity of the modelling process and the possible validity of the constructed model. Central to
these problems are issues of developing appropriate conversion scales as discussed by Nuthman’,
and general managerial scepticism of computer models.
Conceptual Model
The conceptual preparedness model Figure 1: Conceptual Preparedness Model:
adopted by Defence contains two clements
Response to Strategic Warning
illustrated in Fig 1. The first clement is a
z Deployments
oe P 2 lta omtes ,
relationship of required preparedness over Sa]
Es on
time. The peace time level of capability is sct $ ss
. : J | pect 2
at the minimum value consistent with the = & (\\ cme i
Pedlormance e
ability to get to the target operational level of 1 Time 2
and
capability within the scenario time frame. an ha Deployment Tine
\_/ Components of Force
}- ‘Structure
The second element represents the components of force structure; personnel, equipment,
and training. Much of the difficulty in developing effecting resourcing strategies lies in a lack of
quantified understanding of how the components of force structure combine to create a position
on the capability axis. The problem is to quantify them so that policy decisions can be rigorously
tested.
Fundamental Relationships in Preparedness System
This project has set out to develop a general model of preparedness applicable though all
levels and across the components of the Armed Services. Qualitative examination of a range of
these component organisations reveals that the fundamental relationships contained in the
conceptual model are widely applicable but may vary in relative importance and complexity. This
degree of repetition allows us to illustrate and explore the relationships by detailed examination of
a selected component without attempting the enormous task of representing the entire system.
An Army helicopter unit, The 5th Aviation Regiment, was selected as the platform for
thorough study. The reasons for this were:
© In June 1996 two of its aircraft had collided killing 18 aircrew and passengers. “The board (of
inquiry) found a number of long term systemic factors which contributed to the accident...”
¢ Reliance on aircraft availability to conduct any activity forces inclusion of maintenance and
logistic issues.
© Close training relationships with other units requires the model to cross organisational
boundaries and adopt a capability output focus,
¢ The system complexity and multiple roles of the unit allow testing of the relationships in a
highly sensitive case study.
¢ Aviation is the only combat or combat support capability represented in all Services which
allows the Defence Headquarters to consult stakeholders with a commonly understood
framework
Broad Structure of the Model
The components of force structure; personnel, equipment, and training; form the modules
of the preparedness model. Each of these components can be separated for individual validation,
7
but it is the links between them which are critical to this study.
Personnel
Personnel are assumed to exist in one of two states (shown in Fig 2): Active flying duty, or
other duties. Attributes of length of time in the system (which controls promotion and separation
rates) and a scaled measure of skill (derived from rate of effort), are maintained for each intake
cohort as array variables.
Figure 2: Personnel Component
Phir Duties Cuan Duties The personnel management system is
Intake, os ‘ SY - fundamentally geared to long term (peacetime)
7 stability of career structure based on broad
Effort Le fort L te command experience. This objective conflicts
Separation Separation
with the short term horizon of the operations
manager who wants to ensure sufficient numbers of personnel with a skill mix appropriate to
potential short lead time scenarios.
Equipment
The equipment module has a similar structure to personnel except that an additional state is
required for unserviceable equipment. In the case of 5 Aviation Regiment aircraft serviceability,
affected by both the supply of spares and the capacity to perform maintenance, critically affects the
ability to train and maintain crew skills.
Training
The model assumes that all flying activity directed to a given task (eg. Counter terrorist)
contributes in some degree to training effect
Figure 3: act of Activity on Effort
for other roles. This is illustrated in Fig 3,
Required to Deploy
where training for a task may have some
Rate of effort available pales OMS Re
Effort ole 1 ate of Decay
effect on the retained skill in another. Priority
seus es om nase
is given to minimum currency activities such poexton:
as emergency drills, after which general tasks
or role specific training may be conducted. A
a te Effort rote 2 Rate of De
critical relationship is how much the effort Role 2 Workup Required
directed at one role contributes to skill in the
772
others. The capacity to achieve training tasks is limited by crew and aircraft availability.
This module provides the critical performance indicator of time required to deploy. Each
role has a defined set of work up activities required before deployment. The length of time
required for these activities depends the recency of training effort in that task and on the rate at
which skill in that task decays.
The actual time required to conduct these work up tasks, and hence deploy, still depends
on crew and aircraft availability
Impact of Simulation Game
Simulation games have an established place in management development. In most cases,
such as the beer game, there is sufficient commonality between participants’ stations as to allow
fairly simple comparison through a single unit of measure such as money. The simulation game
built from this model has no such luxury.
In this game there are 5 stations, each of which is operating on a different length of
decision cycle, different length delays for the impact of decisions to occur, and different
performance measures and criteria. Players form two broad groups. First are those responsible for
long term policies such as training standards, manning levels, and promotion windows. The second
group are those responsible for resourcing and tasking,
usually within an annual cycle although there may be long Higure 4: Game Interface
Personnel
Resources & Policy
term effects from some decisions.
Figure 4 Illustrates some of the major information ‘Training Policy
flows in the game. Throughout the game each player | Financial Resouces fo
vit
——~as
receives only the information normally available to his = a Tasking
Uni
position in the organisation, and is able to judge his _
Games Master
performance only by that information (eg. manning levels
or task hours achieved). The complex decision events
such as what type of training to conduct are retained by
the players. The model simulates the impact of policy, resource and tasking decisions on aircraft
availability, crew skill, and training levels.
773
At any stage the games master can impose a Defence Scenario which may differ radically in
both lead time and size from the planned responses to contingencies. Success in the game is judged
by the capacity to reach deployment standard in the required time. This can only be achieved if all
players have worked cooperatively during the early part of the game, attempting to understand
how their decisions will impact on force capability. In the current management systems, however,
performance indicators encourage ‘individualistic’ behaviour.
The requirement for cooperation can be illustrated by a sample of the interactions between
the players responsible for personnel and training policies. The personnel agency sets policy
related to the length and spacing of rotations into the unit. Performance is reported back in terms
of manning levels by rank. The training policy agency identifies the experience required for skill
progression and the relationship between flying activity and training effect. Performance is
reported in terms of numbers at each skill.
Because flying skills decay when pilots are posted to corporate duties, personnel policies
impact on training policy decisions, without training policy decision makers being able to adjust
personnel policies. Action to
Figure 5: Interaction Between Training and Personnel
address training deficiency by
Lap Ng foe sul increasing activity will in turn
Acquire Sk YE! sie Decay Acquie seit L2¥*!__ suit Decay . oo
Oo 3f 3c) impact on the serviceability of
Le [
the aircraft,
Time in
Corp Mngt
Initially, the game is used
to validate the —_ difficult
aan relationships in the system by
Personnel Policy Maker | Training Policy Maker
exposing them to the players as
decision parameters. The intent is
for domain experts to fix levels and then discuss the results of the simulation to validate these
relationships. This validation is expected to produce a very robust model of this particular, but
unusual, organisation suitable for use by training and personnel planners.
I would argue that this is exploiting only a small part of the potential of the model. More
senior executives exposed to this simulation are faced with quantified relationships captured in a
774
]
dynamic simulation. The impact on capability of particular decisions on the relative importance of
individual tasks can be tested. Especially interesting is that this impact is tested beyond the
expected tenure of those making the decision,
The use of an aviation unit as the platform for the simulation means that differing
assumptions of the three services are exposed, and a common understanding developed. Thus,
during resource allocation the debate can focus on the relative strategic importance of the specific
role with a clear understanding of the impact over time and the contribution to other roles. More
than this, debate does not need to be preoccupied with understanding the analysis when the
fundamental relationships have been developed in a joint environment.
Conclusion
The peculiar strategic environment of Australia’s Defence Force requires a sound
understanding of the complex interaction between components contributing to preparedness.
Without this understanding the long lags between decision and impact make effective peace time
resource allocation impossible. This study has developed a learning environment which allows the
impact of resource decisions to be explored in a system with no common scales of measure,
competing objectives, and substantial organisational barriers to communication.
Now that the concept is proven, however, the next step identify representative classes into
* which units may fall. A set of models can then be built which would allow decision makers to test
the impact of their decisions and determine the information requirements of key participants for
capability based resource planning.
' Minchin T, Robinson P, Long T (1996). M: mi i fence Force
, in: The Auditor-General Preliminary Study Audit Report No 17 1995-96 .
Canberra Australian Government Publishing Service. p3
? Nuthmann C (1994). Using Human Judgement in Systems Dynamics Models of Social Systems,
in: Systems Dynamics Review Vol. 10,Nol .US John Wiley & Sons, Ltd
3 Ministerial statement from Board of Inquiry 6 Mar 97 www.adfa.oz.awDOD/Minister
775
CC BY-NC-SA 4.0