314
A DYNAMIC MODEL
FOR THE PLANNING OF HUMAN RESOURCES
Habib Sedehi
ENIDATA - TEMA S.p.A.
Viale Aldo Moro, 38
Fiera District
40127 Bologna
Italy
1. ABSTRACT
The methdology which this article deals with, is the
fruit of the experience gathered thanks to the contributions
of experts in the fields of programming, personnel and
organisation.
The willingness of such people has given rise to the
instruments now available in the field of Industrial pDy-
namics, It is thus possible to design a methodology and
simulation models which, with the help of a computer, offer
an instrument which can answer the needs of those involved
in the planning of human resources in a company environment.
This model has so far been applied with satisfactory
results in two technically different realities. At the
present time, other applications are in progress which
confirm. the validity of this instrument. py
1. INTRODUCTION
When drafting company programs and strategies,
particularly in the productive, economic and financial
sectors, one can certainly not ignore human resources.
Notwith-standing this, in the great majority of Italian
companies, both private and public, management has so far
given little attention to personne] planning and has very
rarely shown the ability of treating the problem with
a real quantitative methodology capable of including all
ene aynanice. [2]
It very often happens that there is an “unbalance” in the
number of employees for several reasons: resignations,
retirement, re-training, etc. Confronted with these situa-
tions, management often finds itself unprepared and in
order to find an immediate remedy, risks taking hurried
decisions which, over time, can lead to serious financial
losses, With regard to the Italian situation, it must also
be remembered that the programming of human resources has
been obstructed by a series of rather unfavourable
conditions, such as:
* the demand for a workforce which can easily adapt to the
existing process-product type;
* a progressive legislative rigidity giving high job
security which has obstructed programming processes;
* the homogeneity of the variables to be introduced into
the planning process;
* the structure of personnel information systems(auditing
or cost control).
315
2. .A DYNAMIC MODEL
In this paper we describe an instrument which, by
using the methodology of "Dynamic Models for corporate
Strategies" (MDS, Modelli Dinamici per Strategia aziendale)!)
allows the forecasting of personnel trends based on the
activities of particular sections. [3] [4]
This instrument, complete with an analytical methodology
developed by a working group 7) is 2articularly suitable
for medium/large companies, and may be applied independently
of the sector in which they operate,
Starting from the assumption that the act of programming
is much more a question of designing rather than simply
foreseeing, a Dynamic Model has been developed which allows
1)
The MDS methodology has been developed by ENIDATA-TEMA ,
a company of the ENI group (Italian National Hydrocarbon
Agency), by elaborating and integrating concepts of Industrial
Dynamics. ig]
2)
The working group was composed of: coordinators:Gio-
vanni Parillo (ENI) and Vincenzo Gervasio (Tema); members:
Giampiero Donnini and Piergiorgio Filippucci (ENI), Alberto
Trebeschi, Claudio Nanni and Habib Sedehi (Tema), Felice Ata
nasio (Saipem), Ferruccio Costetti (Anic), Agostino Romoli
and Aldo Rossi (Tic).
the verification of the validities of the different politics
of intervention, by asking the following question: “Given
a certain initial situation (either stationary or varying
with time) and having made several hypotheses of interven-
tion (or non-intervention), how will the workforce change
over time?".
This mathematical model allows the determination of
personnel costs (both for stable employees and those being
trained) as a function of the different hypotheses chosen,
and also an analysis of:
1) increases and/or reduction of the workforce in given areas
of a company;
2) trends of internal and/or external mobilities;
3) attribution of priorities to different personnel categories;
4) the timing and the preferences which govern mobility from
one category to another.
According to the classification made by Bowey
"Corporate Manpower Planning", [s] which was also used by
Buckley, Iantsh and others (although using a different
terminology), in the real operative situation of a company
there are three different possibilities:
A/ equilibrium (in which the organisation is stable over a
considerable period of time);
B/ adaptive (in which the organisation varies due to internal
stimuli);
c/ planned (in which the organisation changes according to
choices made previously).
The experience gained in the use of this methodology
allows one to show how all three approaches are characterised
by results which can not be underestimated and is not,
therefore, subject to limitations which often give to such
a project a purely theoretical value. It was thus necessary
to define a representation of company structure which allowed
the testing of all three hypotheses, or better still, to
verify the effects of all three and their different combi-
nations.
The problem would also have to be approached both in
terms of:
- optimisation (in the sense of minimising the costs over
the interval required for reaching the desired situation);
- and simulation (in order to have a collection of results
as a basis for decision making).
The complexity of the "process" and its peculiar
characteristics governed the choice of a simulation instru-
ment which has contributed, amongst other things, towards
showing the necessity of close links between personnel
planning and all the other company activities. In fact,
personnel planning can not be. considered as a "one-off"
process, but as a dynamic activity, like all company
activities, which require continual adjustment,
Thus, remembering that the interest of Planners is
Normally centred upon forecasting supply, mobilities and
demand, emphasis has been given to the company as a dynamic
entity, with all of its ability to react to internal and
external stress. iD
The resulting instrument, complete with an appropriate
methodology of analysis, is suitable for use in medium/
large companies, independently of the sector in which they
operate.
‘317
3. METHODOLOGY OF ANALYSIS
The methodology of analysis adopted in the definition
of the structure and of the data of the model allows one to
draw together the following aspects of the problem:
* structural analysis of the internal organisation;
* integral analysis of company budgets and strategies;
* analysis of personnel files;
* analysis of the social situation;
* analysis of internal mobilities.
Structural analysis of the internal organisation.
Used to subdivide the personnel in the operative unit under
examination (society, factory, area, division, etc.) into
homogeneous classes of functions having a set of requirements,
or "describers", in common. Let us suppose that we wish to
analyse the functions of a hypothetical company by means of
three types of macro-describers:
a) qualification (ay diploma, aye arts or social science
degree, aye science degree);
b) age group (b,= 20-30, b,= 30-40, b,= 40-60 years old);
c) knowledge of foreign language(s) tee English, c=
French, c3= German).
Let us also assume that in the above company the
functions of welfare officer and public relations officer
can be defined using ay and b)- If we now construct a
describer/function matrix, the intersection of the rows
and columns represents the need for a describer for that
function, and the functions which can be aggregated are
those having the same describers (functions 2 and 4
in Figure 1).
It is, necessary to have a wide and detailed knowledge
of the reality under examination in order to verify the
validity of the results obtained so that errors and over-
simplifications can be avoided,
Otherwise there is the risk of grouping together
positions having the same describers but different functions
or, vice versa, of not grouping together positions having
different describers but similar functions. The sets of
homogeneous positions resulting from this analysis are
called "families".
Internal analysis of company budget and strategy.
After having found the families, the duration of the simula
tion must be decided upon, by taking into account the aims
and goals which the company intends to reach. Towards this
end, the requirements of the company budget are integrated
with the course of the strategic planning (investments,
disinvestments, personnel retraining, technological inno-
vation, etc.).
318
lo.
Public
Secretar Porters relations,
retary be Officer bt.
4 E 3 4
x
* *
oT
0-0 x
e tr
& Ed * *
» v8
F 40-60 x
d
& Inglish x
a
rH french
a
= German
xz
5
Fig. 1
ll.
This analysis may lead to a possible revision of the
previously classified families (for example, re-training
may introduce new functions, and conseguently, different
groupings). :
Analysis of personnel files. This has the following
objectives:
- to find the number of employees belonging to each family
identified (from the personnel information system);
- formulation of hypotheses on probable personnel losses
(natural wastage) due to resignations, illness, etc, (from
an analysis of the history of the structure);
- projection of known personnel ldsses, due to retirement
and other causes (from the personnel information system).
Analysis of the social and trade union situation.
In order to include external effects on the structure it is
necessary to analyse the different variables (schooling,
Population, job market, etc.) and to identify those which
exert the greatest influence on the reality in question,
attempting also to correlate them.
To reach this aim one must construct a behavioural
model which allows an evaluation of the quantity and quality
of the personnel which the market offers. Regarding the
trade union system, this analysis also serves to underscore
constraints on mobility, working hours, different skills,
etc.
12.
Analysis of internal mobility. This step allows one
to bring together the time, costs and limits involved in
moving personnel from one family to another. Por this, one
can construct a mobility matrix with the rows representing
the acceptor families and the columns the donors. ( see
Pig. 2).
If there is a number at the point where a row and a
column intersect, this represents the time required for the
transfer from one family to another, the absence of a number
means that such a transfer is not allowed. For example, to
transfer a person from family 1 to family 3, nine months’
training are required.
Once the different steps in the analysis have been
completed, we have all the structural elements and the
necessary data at our disposal in order to construct the
model (see Figure 3).
a5
Family) Family | tamily | fee
2 3 ri
3 months 6 months
family 2 Ament,
family 3 9 months
Farrily 4 A months
Eke.
Fig.
2
320
--4 >. | Structural analysis of
Integral analysis of budget |-----—-— —-4
and strategic plan
—s|l Ik
14,
internal organisation
Giealattion model
Ll oOo
r
uoqoms von apa
[pur jones fo srsfiouy
Analysis of Analysis of
Personnel files intemal mobility,
15.
4, MODEL STRUCTURE
The model consists of a set of differential equations
which describe the process (deployment of personnel) thus
allowing a dynamic simulation of the system within the given
limits.
From the structural point of view (see Figure 4) the
model is mainly composed of three logical elements:
- levels, the state variables of the system;
- flows, elements which modify the values of the levels over
time
» endogenous calculated by the model
. @xogenous inserted from exterior
- parameters, coefficients used to define the laws of the
endogenous flows.
* Levels
The levels considered are:
- External market; represents the feedstock of the system.
- Persdnnel training; the so-called "waiting room" for
employees which are about to change family.
- Operating personnel; the “operative force" of the personnel
structure.
- Departures; represents the reservoir of personnel leaving
the structure,
321
Firing
16.
Probable
Operative
ersonnel
yi
ao
Fig. 4
certain
Departures
its
* Flows
The flows identified are:
- hiring of personnel from the external market;
- mobility of personnel from the donor family to the formation
of the acceptor family;
- personnel transfer from training to the operative personnel
of the family;
~- personnel resignations (probable departures);
- retirements (certain departures).
* Parameters
The principle parameters adopted are:
- training period required to pass from one family to
another;
- preference for mobility from the donor family to the
acceptor family;
- desired workforce of the families;
+ minimum and maximum percentages of family employees with
respect to the desired workforce;
- percentage distribution of qualifications across the
families;
- operative and formation costs of the families;
- percentages of resignations from the families with respect
to the total.
18,
All the parameters over the entire duration of the
simulation may be subject to programmed variations,
The elements of the model structure (levels, flows
and parameters) may have dimensions (space); for example,
the "operative" level of the personnel is composed of n
sub-levels which corresponds to the number of families
identified in the analysis steps.
Exploding the model (Figure 5) and focussing upon a
family (Fam, a) one can note the possible interactions
existing between such families and the outside world.
The circled numbers on the arrows represent the number
of months' training necessary for a person to pass from the
donor family (or the external market) to the acceptor family
(training time parameter), whereas the uncircled numbers
define the order of preference with which, when necessary,
the acceptor family draws from the donor family.
19,
Fig. 5
323
20.
5. PROCESSING LOGIC
The principal aim of the model is as follows: given
the initial state of the workforce subdivided into "fami-
lies" (initial values of the Levels) and given a desired
state of the workforce accompanied by a series of constraints
(values of the Parameters and their variation with time),
what will be the evolution of the state of the "families"
for a given duration of the simulations?
The model provides two types of simulation for reaching
the objective:
a) Automatic Simulation (quantitative aspect).
b) Guided Simulation.
These two simulations are always preceded by a third(ageing
aspect), which may be carried out independently and repre-
sents the relaxation of the system. In fact, this simulation,
by freezing hiring and internal mobility clearly shows
the "emptying" of the reservoir of the family operatives due
to retirements, resignations, or other departures, and
identifies which families would become inoperative due to
the lack of personnel. This simulation also provides further
indications on the critically important professional figures.
a) Automa’ Simulation (qualitative aspect)
Starting from the initial conditions, fixed on the
basis of the results of the ageing simulation, this simula-
tion determines the families which will become quantitatively
ah
324
more critical thus automatically creating a resolution
priority vector. This vector consequently modifies the
number of allowed transfers since the model,resolving the
families one at a time on the basis of their assigned
priority can not upset the order of an already settled
group. Having done this, it simulates, on the basis of
the automatic priority vector and other parameters of
the model, the evolution of the described levels.
b) Guided Simulation:
Given the initial conditions, by choosing this type
of simulation, it is the user who defines the resolution
priority vector and who also has the possibility of inserting
possible exogenous plans for mobility or for hiring, and
therefore the model simulates the behaviour of the levels
described "guided" by indications from the user.
In these two principle simulations, within the priority
vector, the transfer of personnel from one level to another
takes place by bearing in mind three elements (in the follo
wing order of priority); -
- training period:
- removal of the operatives from the workforce of the donor
family;
- transfer preferences,
22,
6, PRESENTATION OF RESULTS
The system provides different types of output, bothof
a graphic and of a numerical nature, such as:
The mobility plan. This may be: considered as the most
complete output for each family, because, after each simulation,
it furnishes all the movements of the employees: belonging
to the family under examination, both to and from other
families. In particular, it specifies both internal movements
{internal mobility) and external ones (hirings, reti-
rements, resignations, etc.).
The training program. This output gives the trends in
personnel being trained underlining the relative periods,
both individually and globally.
The employment plan. If new appointments are foreseen, then
this prospectus indicates which ones are to be made by
specifying the qualifications required for each case
(the percentages of different qualifications desired for
each family are input data).
Cost_report. For each family and/or globally this output
gives the training and running costs in graph or numerical
form,
All the prospectuses and graphics of the system can
be requested with any desired frequency which can also be
changed at will by the user, within the limits of the
duration of each simulation.
23.
7. CONCLUSIONS
This model has so far been applied with satisfactory
results in two technically different situations. The first
was a chemical plant having about 4000 employees and a
problem of global planning of the workforce. In this case
the main difficulties arose from the structural analysis
of the internal organisation, and in particular from the
identification of the families, whereas the model respon-
ded well to the requirements of the problem,
The second case involved the planning of only the
intermediate levels (about 1000 employees) of a large
company. This application showed fewer difficulties in the
analysis. stage, but required several modifications in order
to correspond more closely to the reality being simulated,
Both cases, however, confirmed the validity of this
decisional aid for the manager who has to design a personnel
structure in a "turbulent" environment such as the present-
day one, and who wishes to test his ability to judge si -
tuations by widening the number of solutions to be taken
into consideration.
325
24.
8, BIBLIOGRAPHY
| H,Sedehi: "Progettazione delle risorse umane con
l'aiuto del computer" DATA MANAGER(1983).
fe G.Petroni: “La pianificazione delle risorse umane".
3] J.W. Forrester: "Industrial Dynamics", New York(N.Y),
John Wiley & Sons Inc. (1961).
( J.W, Forrester: "Principles of Systems", Wright Allen,
Cambridge Mass. (Trad, It. Principi dei
Sistemi, Etas Libri, Milano 1974).
E, Bartezzaghi e S.Mariotti (Editors): "Modelli per
lé decisioni strategiche aziendali, il si-
stema MDS-TEMA (ENI)", Angeli,Milano (1983),
(J A.N, Bowey: "Corporate Manpower Planning".
| A.R, Smith: "Models of Manpower Systems".
