A MULTI-METHODOLOGY APPROACH TO ADDRESSING ICT SKILL
SHORTAGES INA GOVERNMENT ORGANIZATION: INTEGRATION OF
SYSTEM DYNAMICS MODELING AND RISK MANAGEMENT
Dr Alan McLucas and Dr Ed Lewis
University College, University of New South Wales
Australian Defence Force Academy
Northcott Drive, CAMPBELL 2600
AUSTRALIA
Telephone: + 61 2 6268 8332 / +61 2 6268 8178
Facsimile: + 61 2 6268 8581 / + 61 2 6268 8581
Email: a.mclucas@ adfa./edu.au / e.lewis@ adfa.edu.au
Abstract
This paper describes a multi-methodology approach used to reveal feedback dynamics
operating in supply of, and demand for, skilled information and communications technology
(ICT) employees. It describes how a large public service organization has had difficulty in
adapting to rapid technological change, and how this has been exacerbated by employee
shortages. The impacts of changing demographics of the workforce and society are
investigated, along with how these factors impact upon attracting, recruiting, training, and
retaining employees. The paper is as much about what is needed to achieve implementation
success as it is about the effective application of tools and techniques in the development of
strategy. It describes the lessons learnt when a system dynamics model of the supply of ICT
skills was combined with a tool that enabled analysis of the risks that shortages would arise,
and the consequent cost of those shortages to the organization.
Page 1 of 16 Pages
A Multi-methodology Approach to Addressing ICT Skill
Shortages in a Government Organization: Integration of
System Dynamics Modeling and Risk Management
Introduction
People, organizations, institutions and environments shape each other (Sastry, 2001: 377).
When the environment within which an organization operates is continually changing, so the
organization must adapt. Such adaptation must balance demand, supply, willingness to
change and effectiveness (Beer, 1966: 474). In terms of workforce, change might take the
form of adjustments through hiring or laying off staff; re-engineering business processes
(Hammer and Champy, 1994: 80), or outsourcing (Leung, 1997); by acquiring new
knowledge, developing individual and collective skills (Davenport and Prusak, 1998;
Katzenbach and Smith, 1993: 47), by hiring new employees with targeted skill sets,
implementing specialised training programs or by institutionalising the valuable tacit
knowledge (Morey, et al., 2002) compartmentalised within in the organization.
This article examines the on-going challenges faced by an organization in balancing the
supply and demand for skilled information communications technology (ICT) employees.
Achieving this balance would enable the organization to provide more effective support to its
parent Government department. Knowing what the future levels of supply and demand will
be is very difficult (Career Space, 2001; ICT Skills Foresighting Working Group, 2006).
Being one of several organizations in a competitive human resource market, knowing how to
influence supply to ensure that its own demand will be satisfied is even more difficult.
Further, any balance achieved between supply and demand is likely to be temporary, given
that the organization operates in a continuously changing environment. So, once having
achieved a balance, to maintain it becomes the next challenge.
Successful organizations select and train their employees to be skilled and knowledgeable in
the domain in which the organization operates. Such organizations maintain their
preparedness by being sensitive to what is occurring both inside the organization itself and
outside in the environment. Whilst the endogenous focus might be on the number of
employees needed in each of the required employment categories to efficiently meet the
operational demands placed on the organization, the organization must remain cognisant of
exogenous factors such as the capacity of the employment marketplace to respond to
increasing demand. This marketplace is also shaped by the demographics of the society from
which employees ultimately are drawn.
Organizations adapt through a series of change interventions. Each of these takes time, and
several may be enacted at any one time. The success achieved depends on the ability of those
who design and implement the interventions to determine the most appropriate changes.
Given the inertia of Government organizations, the need to change must be established early
and change interventions must be implemented well in advance of the date that results are
expected. Senior executives also need to be convinced of the veracity of the likely future
interventions and be strong enough to lead the organization through the necessary changes.
Page 2 of 16 Pages
For the organization, the following make it difficult to establish and maintain a workforce
best able to deliver the required outcomes:
e the environment within which the organization operates is changing rapidly;
e the nature of the organization’ s business is changing, and at an accelerating rate;
e the demographic profile of the society from which its employees are drawn is weakest
in the required age groups;
¢ potential employees having the desired skills are falling in number; and
e the demographic profile of the organization shows a preponderance of workers in
older age groups.
In this article we focus on how to design truly effective human resource management
strategies. The organization has been experiencing increasing numbers of separations and
growing mismatches between required and extant employee skills. Strategies previously tried
have not produced the desired results and senior executives have appointed a senior group of
middle managers to investigate and recommend effective remedial strategies. The newly
appointed strategy development working group reports to the senior executive steering group.
So, convincing them that any particular strategy is likely to correct the organization's current
and potential future problems will also prove to be a significant challenge.
However, this is a challenge that must be met because obtaining the unreserved support of the
steering group is essential to the subsequent implementing of any management initiative.
They will play an important role in obtaining higher level support, should initiatives need to
be implemented seeking to achieve broader effect, say, in influencing supply. Such a need
might be indicated by the existence of a strong dynamic influences operating in the wider
manpower marketplace, needing expansion of the problem boundary beyond that initially
defined.
Because the organization is part of a government department, departmental or even specific
government support may be needed before strategic options such as those affecting society
might be implemented. The first step agreed by the strategy development working group was
to build a comprehensive understanding of the dynamics and how they work to influence the
demand for, and supply of, skilled personnel in the organization, the government department,
those industries upon which government relies, and society more broadly. This understanding
would prove essential to being able to proceed with their strategy development task.
Possible strategies discussed early by the strategy development working group included
outsourcing as a way of reducing the organization’ s reliance on their own diminishing
numbers of skilled employees. The efficacy of such strategies is now being questioned
because of the risk that outsourcing may not solve the real problem. Consequences might
include shifting undue burden to industry, which industry may not be able to carry or may be
politically unacceptable if industry’ s profit-making opportunities are significantly lessened.
Whilst outsourcing might produce attractive opportunities for a few large companies in the
industry it is unlikely to be attractive to the many smaller companies who also rely on
government business.
Because of the time it will take to build up specialist skills and knowledge, companies who
seek to exploit the outsourcing opportunity will incur significant costs. It is likely that only
the larger companies may be able to incur such costs in the short-term and offset them in the
longer-term. In turn, the ‘shifting the burden’ initiative might well limit the organization’ s
ability to downsize or outsource, because to ramp up their capacities companies operating in
Page 3 of 16 Pages
the support of the government will seek to draw from the same limited pool of potential
employees. Therefore, it seems inevitable that they will be tempted to poach employees from
the organization itself. With such possible consequences in mind, the senior executives have
made it clear that all strategies developed for further consideration must explicitly identify the
inherent risks to the organization and associated costs. Further, implementation of short-term
initiatives is not to threaten long-term organizational goals. The analysis and recommended
options are to be presented in a detailed strategic appreciation, noting that long-term goals are
to be achieved through a program of organizational change interventions.
This article explains how, through intimate engagement of the strategy development working
group in group model building workshops, a series of system dynamics and risk management
models were built to inform the development of a strategic appreciation. This was based upon
discovering the complex dynamics that operate in the supply of, and demand for, skilled
employees.
The Problem
Given the continued and rapid changes occurring in ICT, how does this large Government
organization, whose role it is to deliver ICT business services and new capability projects,
ensure that it has the required numbers of people with requisite skills and knowledge now and
well into the future to fulfil its role effectively?
Organizational Context
The organization is part of a large Government department. It is responsible for supporting
the whole department, ensuring efficient, secure and reliable communications and information
handling capability locally, nationally and internationally. The department, and in turn the
organization, receives strategic guidance and direction from the Federal Government. The
organization’ s senior executives define ICT capability requirements needed to support
corporate governance of the department, both in the short- and long-term. ICT capability
requirements are designed to exploit technological opportunities whilst reducing threats to
effectiveness and reliability of communications and security of information. Consequently,
those in the organization must be skilled in ICT management and knowledgeable in IT,
communications technology, information management as well as developments in each of
these areas.
The organization currently enjoys high priority for funding in the Department’ s budget, but
funds are limited. The organization’s managers are focused on achieving value-for-money in
ICT implementation whilst managing risks to continuous and effective ICT service delivery.
They are also responsible for concurrent development of new ICT capabilities consistent with
the department's strategic goals and expected future operational needs. Enhancements to ICT
capability are delivered through an on-going series of projects. The organization’ s capacity to
conduct business as usual and enhance ICT capability depends on availability of human
resources, skills and expertise in each of the key knowledge areas, state of infrastructure
development (only able to be significantly enhanced as a long-term goal) and availability of a
wide range of physical resources.
The role of the organization has evolved to now being a provider, manager and developer of
comprehensive ‘information age’ ICT services to thousands of desktops. The organization is
Page 4 of 16 Pages
also responsible for protecting communications and information systems to the extent defined
by Government security regulations.
The advent of ‘information-age’ technology has dramatically changed the way the
organization operates, and this has imposed new, and in some instances unique, challenges
upon the workforce. The workforce has been forced to adapt to an ongoing series of
organizational transformations as well as seemingly endless and increasingly rapid
technological change. Further demands have been imposed by growing diversity in means of
communication used by the department, need for knowledge of legacy ICT systems, those
bespoke systems not easily paid off or replaced because of their unique functionality, and the
growing and largely unfulfilled need for integration of new and legacy systems.
All these demands have forced an ageing workforce to confront the need to adapt, or be
overwhelmed by rapid changes. With consequent losses from the workforce in prospect it is
becoming increasingly difficult for the organization to deliver to meet its changing role. The
organization is experiencing shortfalls measured in hundreds of personnel in several key
functional areas. Current trends suggest that shortages will continue to worsen.
Recently, a series of senior executive initiatives have been directed at transforming the
organization by re-engineering business processes. These involve a shift from in-house
design, construction, integration and management of IT, communications and information
systems to outsourcing such activities. In the long term the senior executive steering group
aims to retain in-house the critical role of managing projects, which are to be delivered by
certified professional service provider (PSP) companies. The organization’ s aim is to
progressively outsource, leaving it to focus on attracting, recruiting, training, employing and
retaining people in roles of operating, planning, managing and supervising the maintenance
and upgrading of ICT systems. The strategy development working group is disturbed that
behind these initiatives is a set of untested assumptions that the PSP companies, and industry,
generally, have or can develop the capacity and expertise to deliver the outsourced support
needed by the organization.
Possible Consequences of Generic Strategies Suggested by the Senior
Executives
Concems have been raised about the unintended likely consequences of concurrently re-
engineering business processes and outsourcing. These initiatives could create the
undesirable dynamics summarised by the systems thinking archetypes: ‘shifting the burden’,
Figure 1 and ‘tragedy of the commons’, Figure 2.
Page 5 of 16 Pages
Symptomatic __
+ solution
[Delay _|
2 : ‘
Problem
R
synpien Side
HKe
oe)!
solution
[Delay |
Figure 1. ‘Shifting the Burden to Industry’ - Possible C onsequences of Outsourcing
(adapted after Systems Thinking Organization, 2004(a))
As depicted in Figure 1, the problem symptom is that there are insufficient people in the
organization to do the required ICT work. Current practice is to fill the shortfall by
employing professional service providers (PSPs) on short-term cost-plus, fixed-fee or time-
and-materials contracts, which is expensive. It would appear that one way of reducing costs is
to outsource to industry and establish firm fixed-price contracts for deliverables. Outsourcing
to industry costs industry more in the short term, as companies in the industry invest in
developing expertise and capacity. Costs increase and industry seeks to draw more heavily on
the available pool of skilled potential employees in the marketplace. Talented employees
currently in the organization are attracted by industry, which reduces the numbers available to
the organization. Alternatively, as the costs to industry increase, the number of companies
who are willing to increase their capacity to take on such work will decrease, thereby
reducing the number of companies available to take on the necessary work. This tends to
shift the burden back to the organization, without providing the intended fundamental
solution.
/ S ee
», 's results Resource
limit
\ ae
Total = available
activity
7
i “a
st
B's LE)
Figure 2. ‘Tragedy of the Commons’ - Possible Consequences of a Limited Resource Pool
(adapted after Systems Thinking Organization, 2004(b))
Page 6 of 16 Pages
The situation depicted in Figure 2 is that industry ‘A’ takes on more work in an attempt to
exploit a business opportunity created by an apparent increase the total activity. Both
industry ‘A’ and the organization ‘B’ are attempting to increase their activity rates in selected
technical application areas, though initially ‘A’ would be increasing it more. This would
occur as the burden is shifted to industry as depicted by the ‘shifting the burden’ systems
archetype, to which this dynamic is also linked. Industry ‘A’ will attempt to recruit greater
numbers of skilled personnel. Unfortunately, because there is a finite limit to the number of
employees working in this field personnel are attracted away from the organization ‘B’,
thereby reducing its activity rate.
Unsustainable competition is likely to arise between the organization itself, PSP companies,
industry broadly and other government departments for limited numbers of appropriately
skilled potential employees in the human resource marketplace. Because these dynamics are
inextricably linked the shortages of skilled personnel in appropriate age groups will be
exacerbated. This applies particularly those who currently have, or are able to obtain, security
clearances needed for employment in the organization. Whilst Figures 1 and 2 summarise
thoughts about feedback mechanisms and possible consequences they only tell part of the
story and are insufficient for detailed strategy development.
Possible Consequences of Specific Strategies Already Initiated by the
Senior Executives
Re-engineering business processes already initiated by senior executives will impose dramatic
changes, such as shifts in the roles of software programmers or system administrators into
roles as managers of complex ICT acquisition projects. Incumbents may not be particularly
well suited for such new roles, nor be enthusiastic about making the transformations the
organizational change is about to impose upon them. Despite this, the senior executives
remain optimistic about the workforce’s capacity and willingness to adapt to their new roles.
Dynamic Influences on Supply and Demand of Employees
Supply sufficient to meet demand for skilled employees on the job is not guaranteed simply
by the existence of a large pool of potential employees. The process through which they
become employees can have a strong influence on the rate at which demand can he satisfied.
The length of time from a change in demand being registered to being satisfied is affected by,
time taken to complete education and training, time needed to obtain security clearances, and
time taken to become proficient. The adjustment time (Sterman, 2000: 276-80), that is the
measure of the inherent ability of the employment marketplace to respond to change, is
measured in years rather than months.
For potential employees seeking to enter the organization, the requirements for obtaining
security clearances are stringent. This arises because of the special need to have detailed
knowledge of how security systems for communications and information systems, and IT are
designed and implemented. Note that those in industry employed to deliver the outsourced
work require the same security clearances as their counterparts in the organization. The
processes culminating in the granting of a security clearance routinely takes several months to
complete, depending on the level of clearance required. Hence, the delays involved in
obtaining security clearances are similar for both Government and industry employees. Only
the government can investigate individuals’ suitability for granting of a security clearance and
PSP companies must also pay for this service. Further, the cost to PSP companies of
Page 7 of 16 Pages
obtaining security clearances for their employees may not be fully appreciated by those
companies wishing to exploit outsourcing opportunities. Any strategy to outsource to
industry must be implemented with these constraints in mind.
Human resource managers strive to achieve a balance between the needs of the organization
itself and the need to nurture the employees. This nurturing creates favourable working
conditions and employee benefits such as compensation (remuneration), superannuation and
leave entitlements, all of which encourage employee retention. Unfortunately, they do not
guarantee it. For example, a government-funded superannuation scheme designed decades
previously provides maximum benefits to employees at a nominal retirement age of, say, 60
years old. This will not help a Govemment organization, which has a high proportion of
employees in the 55-60 years age group, to stem the tide of retirements over the next five
years. However, the rate of retirements might be reduced by a change to maximum benefits
in favour of those who stay in employment until age 65 years (though, for the organization
this may actually be deferring an inevitable shortage of skilled labour). Further, recruitment
action designed to compensate for these impending retirements is unlikely to succeed if the
demographic profile of the broader population (from which employees would be drawn) is
similar to the target organization (with relatively few youngsters) or if working in IT or 1S
fields is unattractive to youngsters in the broader population. Managing such issues is beyond
the organization without intervention by the Government. So there is a need to build a strong
case upon which representation to Government might be based.
Creating a Shared Systems View of Influences on Employee Supply
and Demand
The interactions between forces of supply and demand, employee expectations, and
organizational needs produce a complex set of dynamics. To understand these dynamics, the
first round of workshops focused on determining what influences the availability of personnel
in the changing environment described. For those who deal most closely with employees
without having to differentiate between the skills associated with each of the specific
employment categories, the workforce problem is basically one of supply and demand, where
the organization draws its employees from an external pool which is defined by its size,
location, the extent to which it is dispersed, its demographic profile, willingness to seek to
work and suitability of individuals seeking work.
From a perspective within the organization there are two main activities; maintaining the
organization’ s capacity to deliver business-as usual, whilst developing the capacity to deliver
capability projects. The background level of business-as-usual is only one of the key drivers
of demand. Because there is an ongoing program of projects, mostly involving the roll-out of
new ICT capability, employees with diverse sets of skills, competence and experience are also
needed. The number of projects being developed or rolled out depends on the level of
capability desired and the rate of technological change in ICT. As technological change is
occurring, many old skills and much knowledge need to be replaced by new. High levels of
operational availability of ICT are achieved through having appropriate skills, competence
and experience in the current workforce to get the job done right, that is, to deliver reliable
ICT.
Demonstration of the organization’s ability to “get the right job done right” increases the
department's confidence in the delivery of new capability projects. In tum, this releases
funding to roll out new projects. This has the effect of creating higher demand to raise the
Page 8 of 16 Pages
levels of desired capability the organization might deliver, and as new project teams are
established. Decisions to increase the size and structure of the organization are likely to
follow. This influences the need to recruit and conduct training programs to raise the levels
of skill amongst those newly recruited. This produces a greater supply of personnel to meet
demand, though not necessarily to satisfy demand fully. How dynamic mechanisms work to
create supply in response to increasing demand came as no surprise. However, it was a
surprise to discover that increasing levels of availability of skilled employees, that is,
increasing supply, increased the demand for more personnel. Having demonstrated capability
to undertake specialist projects leads to obtaining approval to undertake more projects.
The influences on supply and demand identified during the first series of workshops are
depicted in Figure 3. This diagram summarises the shared understanding that had been
developed. It was particularly useful for identifying the leverage points (Senge, 1990: 64) or
pressure points (Coyle, 1996: 222) where effort might be applied to manage the main risks to
both supply and demand.
Demographics
Caeser
4)
Other Organisations
Retention
Training
‘Administrative Process
Delays
HR Policy
Organisational
Structural Decisions
Iraj & N
1 eng £ _
\ 5 ote Soe Doctrine
‘Security Clearances ~ Shr lie “ae S (Allocation .
Other N 80, l i
"hey t \
!
! F
Management of ‘Get right job done
SUPPLY OF PERS risks affecting both | DEMAND FOR PERS right"
A supply and Ss i
usiness demand $ g
Troster (ss i oe a?
‘Aim to deliver See ae
dynamically Sos ge
(3) ee =
% 23,
% se
% 28
Replacemend Projects ae
of Skills Bs
%
- ccholigy Sustainment
Change
Business AS
Map: 7 Usual
Professions 2
Assume profile of Competency / Skill Desired
Selection Criteria Capability
Influences on Supply and Demand of Personnel
Figure 3. Influences on Supply and Demand of Personnel
This conceptual model provides the basis for conducting subsequent analysis, and was
particularly important for examining implications of how the system boundary was to be
defined, for example:
The ability of the organization to attract new recruits would be dependent upon the
demographics of the society. It was considered impractical for the organization to
change the demographics of the society from which recruits were drawn. Rate of
supply of individuals from society was considered initially to be exogenous, though
Page 9 of 16 Pages
later it became clear that this had to be treated as an endogenous variable and the
system boundary re-defined accordingly. This will be discussed again later.
e Being a technology user rather than a developer, the organization relies on diffusion of
technology. The rate of take-up new technology depends on what the technology is
seen to offer to increase ICT capability. Technology diffusion is considered to be
exogenous.
e Recruiting from industry, both local and overseas, labelled as ‘business transfer’ might
provide opportunities to increase supply of personnel having selected skills and
knowledge. The administrative costs associated with achieving business transfers
would be substantial.
Recruiting rates might be increased by making the organization more attractive or by
changing human resource management policy to enable the specific ‘head hunting’ from
industry of personnel having skills in managing ICT acquisition projects. Even at this early
stage, it was evident that to increase supply would require seeking to draw on diverse sources
(see Figure 4). Associated with this are the needs to attract increased numbers of potential
employees, make the recruiting process more effective, and invest in training recruits with the
aim of developing their sense of belonging and loyalty to the organization.
Labour Force -
indigenous
Labour Force -
Intemational
Entering qualified - from University /
VET / Vocational Short Courses
Organisation -
Business
Elsewhere
Pool of IT qualified
people
—
Train - Organisation
Specific
Sources of Supply of Labour for IT Sector
Figure 4. Sources of Supply of Labour for ICT Sector
Assuming that the need to recruit could be established in quantitative terms and by
employment category, recruiting initiatives would not immediately translate into increased
supply because of delays including those for the obtaining of security clearances. This is
depicted in Figure 3 as the ‘Administrative Process Delays’ intervening between
‘Recruitment’ and ‘Supply of Personnel’.
For the target organization, conflicts have often arisen between short-term and long-term
objectives. Short term objectives include having employees available and trained (preferably
fully competent) to do the work that must be done today, this week or this month. Long-term
objectives emphasise grooming employees for succession to higher levels within an
Page 10 of 16 Pages
employment family group. Promotion is necessary for the organization to retain experience-
based knowledge. However, with growing diversity in ICT knowledge areas, progression
paths and promotion within employment family groupings have become ill-defined, unlike in
the administrative careers stream. This lack of definition creates particular problems for
human resource managers who are no longer able to design or work with simple succession
plans for the management of ICT employees’ careers.
Integrated System Dynamics Modeling and Risk Management Approach
System dynamics modeling has been used quite extensively in succession planning to
simulate the dynamics of supply. Succession planning models usually analyse the recruiting
rates to be achieved and attrition rates that might be accommodated to ensure desired numbers
of employees are promoted and available when needed, with demand for employees as an
exogenous force. Such models are based on business rules involving time-in-rank as a
surrogate metric for competence at current employment level, length-of-service as a surrogate
metric for experience overall, and availability of vacancies for successive promotions to
occur. The modeling approach taken here would extend these previous approaches by also
taking into account the demographic profiles of the organization and the pool from which new
employees were recruited as well as risks associated with discrepancies between targets and
actual numbers of employees and the costs of these discrepancies to the organization. A
range of exogenous and endogenous forces, described in Figures 3 and 4, would also be taken
into account.
An integrated dynamic feedback and risk management model was developed. It had three
linked component parts:
e A system dynamics model developed using Powersim™ Studio.
e Analysis of networked Links (AnnL™ ) software programmed using Microsoft™
Excel.
e Anexpansive dataset in Microsoft™ Excel. This captured attributes of the workforce,
such as numbers of employees, age demographics, employment family, extant skills,
time-in-rank, and expected length of service before separation.
The AnnL™ software is designed to prepare business cases. Here, it used inputs from the
supply model and estimates of demand for projects that contribute to business capability. The
gap between supply and demand was treated as a risk, with an impact determined by the
shortfalls in capability that drove the demand. AnnL™ also includes a cost model, which
enables possible options to be assessed according to their value for money. AnnL™ can be
used to examine the risk-adjusted cost of options, changing recruitment policy or the effect
delayed demand. The effects of these options upon expected recruitment rates or demand
figures over time were fed back into the Powersim™ Studio model, as part of the evaluation
of the risks associated with the options.
The system dynamics model exploited the functionality of a two-dimensional ageing array
module:
e Asneeded, similar modules were linked:
0 inchains to representing multiple periods of ageing with promotion, or
o in branches to representing the paths taken in transferring from one family
employment group to another.
e Each row was used to represent a typical individual in an employment family group.
Page 11 of 16 Pages
e Rows were allocated on a pro-rata basis depending on how many typical individuals
were in each employment family group. Because each row was unique the maximum
number of rows had to be specified from the outset and replicated in each connected
module.
e Ages and time-in-rank ageing were based on the typical date of birth specified for
each individual, represented by a particular row. Age, time-in-rank and expected time
before separation were calculated each simulation timestep.
e The required characteristic ageing time was selectable for each row, by simply
inserting a typical elapsed time in the dataset. Ageing was achieved by indexing the
contents of each array element one column to the right each simulation timestep.
Once the required number of steps had been taken, the contents of the row were lost,
promoted into another module or translated into a new module to represent a change in
employment.
e Business rules were designed to allow individuals to pass through various control
gates such as having been recruited and having a need to obtain the lowest level of
security clearance, the individual was allowed to pass after a random time between a
and b weeks, higher levels p and q weeks, and highest level x and y weeks.
e Attrition occurred by the deletion of the elements in the randomly selected row. Once
attrition had occurred the individual was lost from the employment family group.
Rows were selected and deleted on the basis of historical information from which
probability of attrition occurring was inferred. Alternatively, the probability was
estimated by those in the organization having expert knowledge.
e Individuals recruited laterally joined the selected employment family group by placing
them in a new row.
Simulations focused on selected employment family groups for current numbers and likely
future numbers in the organization. These baseline simulations focused on recruiting to fill
known and expected vacancies. They set out to identify the numbers that would need to be
recruited to satisfy demand. Figure 5 shows the typical pattern of growth in numbers entering
the organization as a result of recruiting. Important to the organization is the number of
employees in each employment family group who have become effective, or proficient. The
upper line indicates the target, whilst the lowest line indicates numbers of effective
employees.
Numbers of Attrition during the
Personnel recruiting process
Recruited
Attrition during the
effectiveness-gaining
process
Recruiting
[pF
Time to obtain a
| | Recruited personnel
security ___| n
clearance ‘ effective on the job
Time
Time to become effective
Page 12 of 16 Pages
Figure 5. Fundamental Relationships Between Recruiting (Supply) and Being Effective
(Satisfying Demand)
The vertical interval between the upper and lowest lines represents the discrepancy between
the target and actual numbers of effective employees in the selected employment family
group.
The demographic profile of those currently working in the organization was introduced. As
individual employees aged and approached their retirement ages or became tempted to seek
alternate employment, it became increasingly likely that they would separate from the
organization (see Figure 6). This created growing discrepancies, which had to be corrected by
subsequent recruiting activity.
Success in recruiting would be dependent upon the demographic profile of potential
employees in the market place. From a modeling perspective taking the demographic profile
of potential employees into account only required the model to be extended, and would use
the same techniques developed already. At this point, the inadequacies of the data available
from the Australian Bureau of Statistics were realised. Until specific data relating to the
employment family groups defined in this study could be obtained (the available datasets did
not align across the required range of employment family groups), the modeling effort could
not be taken further. The alternative of making estimates based on expert judgment alone was
determined to be too risky to pursue at this time.
4 employees Target Number of
Employees
iy led PI =A
Current Number of
Employees
Current Number of a
Original Employees
100
I
|
|
I
I
Startof |
Period of
I
I
|
I
Interest
50
Original Employees
Promoted Employees Leaving
Current R ecruiting
Target
Recruited —_ y eee
re flo {71 re |
08 09 10 11 12 13 14 15
Time (years)
Figure 6. Simulation Results - Recruiting to Replace Expected Separations
Page 13 of 16 Pages
Simulations of the system dynamics model identified discrepancies between target numbers
of employees in each employment category and likely actual numbers. Discrepancies were
seen as risks to the organization’ s ability to deliver either business-as-usual or new ICT
capability projects. At the end of each simulated year, the discrepancies identified were
provided as an input into the AnnL™ model. Based on the perceived risk associated with the
size of each discrepancy a cost was calculated. In effect this figure was the cost to the
organization of carrying vacancies and not being able to do the necessary work, which was
the basis for the demand figures also entered into AnnL™ . In tum, this cost informed the
priority for recruiting. This approach would prove to be an important feedback mechanism
which, when analysed through subsequent simulations, could be used by managers to assist in
establishing new policies for recruiting, for example:
e Because recruiting targets are infrequently met in full, there is a strong tendency for
any discrepancy in numbers to grow. This is a classic undershoot dynamic (Sterman,
2000: 114). One possible alternative would be to seek to recruit greater numbers in
the next round to offset the shortfall in the current and previous rounds. However,
compensating for undershoot by recruiting excess numbers would only be permitted in
extenuating circumstances. Government policy normally allows for recruiting only to
fill existing vacancies. An argument could now be formulated to the effect that there
was a significant risk in continued undershooting of the recruiting targets and the costs
associated with this warranted an exception to the policy. If approved, excessive
recruiting was more likely to retum the required numbers of recruits, provided that
sufficient supply existed. Note that in the simulation depicted in Figure 6, to achieve
numbers of current employees near the target it was necessary to recruit at rates
significantly greater than 100% of current vacancies.
e Because of the time taken to recruit, obtain security clearances, become effective on
the job, and accommodate attrition along the way (see Figures 5 and 6), it would be
necessary to find ways to reduce delays and decrease the service times, particularly in
the security clearance queues. The particular queue a recruit entered for the purpose
of obtaining a security clearance would be determined by the employment category in
which they might find employment, and how well they had been prepared by their
vocational education and training. The service time (time to process them) would be
determined by their ethnic background and length of citizenship. Whether service was
completed or not would be determined by the patience of the recruit, the continued
attractiveness of employment in the organization and the attractiveness of other offers
of employment they might receive from elsewhere whilst waiting in the queue. In
other words, the pipeline into the employment pool was very ‘leaky’, with the extent
of the leaks dependent upon several variables identified in the model. The queue
length, service times and attrition rates would be identified later as critical constraints
upon the supply rates that might be achievable.
The Next Steps - Testing the Models
Interventions to satisfy forecast demand need to be designed by taking into account:
e the mechanisms through which recruits are attracted,
how recruits are selected and processed administratively,
the mechanisms used to obtain security clearances,
training delays,
delays in developing necessary levels of proficiency,
risks that discrepancies would arise between demand and supply, and
costs to the organization of having these discrepancies.
Page 14 of 16 Pages
The senior executives had already specified the target numbers of employees in each
employment category to be achieved through business process re-engineering and
outsourcing. The models described above provide the basis for investigating potential
strategies for achieving those numbers. Before being able to test the models it was essential
to refine the preliminary requirements for the model and develop a specification against
which the model would be tested. Note that this specification could not have been developed
until the problem had been investigated in some detail. Iteration is very often necessary when
investigating complex problems (Homer, 1996). This specification would also enable the data
collection needs to be identified. The specification covered:
e the current numbers of employees in each employment category;
e the target numbers of employees in each employment category for each of the next
five years;
e the probability of losses through separation occurring over the period, including:
0. transfer to other government departments;
0. resignation;
Oo matemity leave; and
0 death, disability or illness.
e estimates of the likelihood that individuals would separate for reasons of being
unhappy about the role changes that the organization was now attempting to impose
upon them.
Conclusion: Some Lessons from Modeling
The earliest models clearly indicated that there were three key parameters: the time to
complete the recruitment process, time to obtain security clearances, and the time to become
fully competent on the job. These models also suggested that managing retention of
employees having specific skills the organization wants to retain, would be critically
important. Finding ways to increase rates of lateral recruitment would also be important.
The working group’s need to simulate various scenarios to answer “what if?” questions about
how the required mix of skills needed by the organization could be achieved, raised issues
about how the models should be designed. Selection of the most appropriate level of
aggregation in the treatment of employees in the model and characterization of individuals
within employment groups became important considerations. Having made particular choices
would mean that the design of the model would be subsequently difficult to change. Design
choices had to be carefully considered, and based on the working group’s needs.
Achieving a suitable mix of skills in the organization led to asking whether recruitment policy
should be flexible, that is, allowing for ad hoc recruitment - hiring in anticipation of demands
project by project, establishing a minimum annual recruitment rate - such as a having an
annual intake of university graduates, recruiting to replace those lost, recruiting to replace
those lost plus those that might be lost before becoming effective on the job, or a mix of these.
The simulations and scenario planning identified other alternatives, including recruiting
individuals pending ultimate granting of security clearances accompanied by a deliberate
choice to invest more heavily in in-house training of those recruited.
Page 15 of 16 Pages
Whilst it may be tempting to make working in the organization more attractive to potential
recruits by offering them more favourable employment benefits, such initiatives would
increase costs to the organization. These additional costs would then impact negatively on
essential organizational activities by reducing funds available in the organization’ s budget.
The associated risk would be acceptable provided the costs were relatively low. If costs were
to increase disproportionately for small or moderate increases in risk, then the option would
become unattractive. In effect, integrating the two modeling approaches captured previously
ignored but important feedback mechanisms involving cost and risk in the context of the
whole of the organization’ s activities.
The integrated modeling approach provided greater insights than one built in just one tool or
the other (or involving one exclusive modeling approach). Group model building proved to be
most useful in informing strategic decision-makers of the implications of their proposals.
There was considerable change in the problem-as-first-stated and the problem-as-last- stated.
Only when conceptual system dynamics models were developed did the true systemic nature
of the problem begin to emerge, and acceptance by the key stakeholders begin to develop.
Members of the strategy development working group involved in the workshops came to
appreciate how each of the feedback dynamics operated and how these produced the
symptoms they had been observing, but previously had been at a loss to explain.
The combining a systems dynamics and risk management modeling approaches enabled the
surfacing of feedback mechanisms unlikely to be revealed through the separate use of these or
other methodologies. Understanding of these feedback mechanisms enabled the development
of several strategy recommendations likely to be much more effective than those previously
attempted. Specific recommendations regarding the gathering of data about variables which
influence supply and demand will be crucial in the testing and development of subsequent
models. A sense of shared ownership of the strategic recommendations, expressed by the
members of the working group, stemmed from being involved in the group model building
activities and experimenting with scenarios developed by the working group.
Despite being unable to take the modeling effort to its logical conclusion because of lack of
data, sufficiently powerful insights were developed to enable the strategy development
working group to argue strongly for the recommendations contained in its strategic
appreciation report. They are now confident that they can convince the senior executive
steering group of the efficacy of their strategy recommendations.
References:
Beer, S., 1996, ‘Decision and Control: The Meaning of Operational Research and
Management Cybernetics’, Wiley.
Davenport, T. and Prusak, L. (1998), ‘Working knowledge’, Harvard Business School Press,
Cambridge, MA.
Career Space (2001), ‘Determining the future demand for ICT skills in Europe’, www.career-
space.com/downloads/supplydemand.doc. (last read 21 Feb 07)
Coyle, R.G., 1996, ‘System Dynamics Modelling: A Practical Approach’, Wiley.
Homer, J., 1996, ‘Why we iterate: Scientific modeling in theory and practice.’ In: System
Dynamics Review, 12(1), 1-19.
Hammer, M. and Champy, J., 1994, ‘Reengineering the Corporation: A Manifesto for
Business Revolution’, Harper-Collins.
Page 16 of 16 Pages
ICT Skills Foresighting Working Group (2006), ‘Building Australian ICT Skills’,
Commonwealth of Australia, Canberra, Australia.
Katzenbach, J. and Smith, D.K., 1993, ‘The Wisdom of Teams: Creating the High-
Performance Organization’, Harvard Business School Press.
Leung, A. (1997), ‘Human Costs Modelling for Policy Leverage in Organisational re-
engineering’, unpublished PhD thesis, University of UNSW, Australian Defence Force
Academy, Canberra, Australia
Morey, D., Maybury, M. and Thuraisingham, B., (eds) 2002, ‘Knowledge Management:
Classic and Contemporary Works’, MIT Press.
Sastri, M.A., 2001, ‘Understanding dynamic complexity in organizational evolution: A
system dynamics approach’, in: Lomi, A. and Larson, E.R. (eds), ‘Dynamics of
Organizations: Computational Modelling and Organizational Theories’, AAAI Press.
Senge, P., 1990, ‘The Fifth Discipline: The Art and Practice of the Learning Organization,
Doubleday, New Y ork.
Sterman, J.D., 2000, ‘Business Dynamics: Systems Thinking and Modeling for a Complex
World, Irwin McGraw-Hill.
Systems Thinking, 2004(a), ‘A journey in the realm of systems’ - Shifting the Burden
Archetype, at: http://www.systems-thinking.org/arch/arch.htm#archtc
Systems Thinking Organization, 2004(b), ‘A journey in the realm of systems’ - Tragedy of
the Commons Archetype, at: http://www.systems-thinking.org/arch/arch.htm#archtc
