The 36" International Conference of the System Dynamics Society
August 6 — 10, 2018 — Reykjavik, Iceland
A Hybrid Modeling and Simulation
Approach for the Strategic
Workforce Planning Problem
Joachim Block
archibald@ieee.org
der Bundeswehr } ] E E [E ®
Universitat (9 Miinchen member of tennsecomin | IY
Humanity
Strategic Workforce Planning (in its basic form)
A14 A15 A16
A13
recruitment / >~ promotion / promotion a ~ promotion >
¢ — 140 193 147 A
wy, he NY KY
A13
[_] Organizational grade = = wastage in
@) Maximum number of
employees in grade
ED Environment > i> o> o>
Modeling a workforce system concerns three types of flows, namely the
recruitment flows, the internal personnel flows between the different
personnel categories (among others promotion flows), and the wastage.
der Bundeswehr
Universitat R Miinchen [GUER2011] 2
Strategic Workforce Planning (in its basic form)
Associate Professors
300
Associate | | Full Professors 2
Professors Professors a ”
Assistant Hire Associate S iso
Rate Promotion Rate 3g i
ae _
5
<Time> Nu Full Professor
DX] ,, Associate Retirement Rate 0
Assistant Departure Rate 19301942 —+1954 1966 —<‘OTS 19902002
Historic Assistant Departure Rate ccietant v Time (Yea!)
Hire late YR_ Review Fate x Review Rate Associate Professors : current
Average _
Average Assistant Associate Review
Review Time
Average Full Professor
Service Time
.
Assistant Promotion ; ; h | t t
ae secaoconotn WWhat are the lever points to
Historic Assistant Historic Associate
Promotion i
a 2% Fangs A, control the system
der Bundeswehr ctihes [STER2000] <Time>
Universitat RR Madnchen “” (into which direction)?
Strategic Workforce Planning (revisited)
HR policies and practices, external environment
Designing a high y y gy
performance
woikagerens, 209 A14 A16
A15
recruitment / > promotion / > promotion Y > promotion Ys
¢ >» 140 193 147 44
wa, wa, wa, KS
A13
wastage i wastage wastage
im Organizational grade wastage
@) Maximum number of
employees in grade
€> Environment o> i>
J
T
\
KPl’s: turnover, vacancies, demography, performance, ... ,
Modeling the Complex Adaptive System
workforce ~ os -
system — * Cp {40} — ‘© — (1-7) ~ ©. General rules
“i = '- aa wastage e— of the system
Exploiting strengths cone CO OG G me pene
by overcoming | | i.
|
barriers ics A> 4 > TP Patterns of
v 5
(ABMS) , behavior
J
an employee
internal decision a Individual
making behavior
(SD + DES)
der Bundeswehr
Universitat RR Miinchen 5
Behavior of an Employee
Ability A Motivation M Opportunity O
o)
Performance P
P=f(A,M,0O)
Universitat rey Miinchen [APPE2000], [JIAN2012] ©
The Hybrid Model of an Agent (an employee)
—>(A \Y>
Based on ora internal event eaiur
well promotion DES subsystem quit
established SD subsystem
DEV&DESS
formalism L798 7 FE \/
continuous 2 weet 9 ON eae Cre oo : continuous
input f fi : to output
HR policies, | fad performance,
overall state IN oe [ \ age etc.
> @) area Vv) =>
der Bundeswehr Steen
Universitat RR Miinchen
[ZEIG2000], [BLOC2014], [BLOC2016]
Running the Simulation (Promotion by Seniority)
Average performance/productivity per employee
130 e. Potential of employee productivity 7
as om
ca Promotion by seniority 07s f= ys <
=: ‘a 7 hy
= 10 4 (A ‘ Ss
g . eh | Ler 7 ag ™ 3
gs 100 4 = ait? ane bs
£ 065 ~ ae =
g ty ©
e 4 06 Zz
05 06 07 0B 09 10 il 12 «13 m4 15 oO —
80 4 ime (yeux) Pad SN
a O
70 zm
0 40 80 120 160 200 240 280 320 360 400 440 480
- complete different approach, but very similar patterns
™—> - SD model limited to few KPI (e.g. no demography) and policies
- attrition fraction has to be set by user (fixed) in SD model
der Bundeswehr
Universitat | Minchen No Calibration to SD Model ! 8
Running the Simulation (Different Policies)
130
Some important results:
— different HR policies reveal
different results
— atraining policy has to be 0]
flanked with other policies
Performance (%)
0 40 80 120 160 200 240 280 320 360 400 440 480
Month
@ years seniority seniority + training performance performance + training
° 1-5 22 24 39 42
= 6-10 24 53 36 87
p> ili-15 4 8 15 23
= 16 - 20 0 0 15 59
m 21-25 0 0 12 Al
S 26-30 ) 37 16 3
3D 31-35 8 30 36 102
O 36-40 8 18 22 54
der Bundeswehr >
Universitat RR Miinchen Sum 66 10 191 ail 9
Evaluation of the Hybrid Approach
-.08* 7 Model
Skill-Enhancing ° . .
. Human Voluntar validation IS
HR Practices \ Capital —10** _s) Turnover
R’= 34 Rs15
re ; a challenge
58** —.28* —.05**
Motivation- _s High Financial
Enhancing HR -_ Performance .21** Outcomes
Practices Work Systems Fl =.27
.62**
.03 37%?
.64** \
Sonoma Employes \ Goouitfonal Simulation results
Buhshiing HR Motivation -16** —3) Outcomes
2 4 2 99
Practices Bhs 38 Rs 22 resemble
iga** % . . .
* Standardized coefficients are presented; n = 3,714. e m pl r ! ca | stud ! e s
a oo JIAN2012] (with one exception)
der Bundeswehr
Universitat rey Miinchen 10
Evaluation of the Hybrid Approach
130 4
120 40 0
Performance (%)
80
70 +
Universitat & Miinchen
110 4
100 4
90 4
¢ different perspectives Vv vv
* more realism ¥ ¥
* reduced model complexity ”
¢ facilitated interaction with r
stakeholders
0 40 80 120 160 200 240 280 320 360 400 440 480
Month
Predictive and explanatory power
Main disadvantages: data source
* multi method skills necessary
[ADAM2017], [Santa Fe Institute 2017 — What are the limits of scientific understanding?]
der Bundeswehr
Selected References
[ADAM2017] Adam C. and Gaudon B.; 2017; BDI agents in social simulations: a survey; Knowledge Engineering Review, Vol.
31, No. 3, pp. 207-238.
[APPE2000] Appelbaum E., BaileyT., Berg P., and Kalleberg A. L.; 2000; Manufacturing Advantage; Cornell University Press,
Ithaca, NY, USA.
[BLOC2014] Block J. and Pickl S.; 2014; The Mystery of Job Performance: ASystem Dynamics Model of Human Behavior, 32"
International Conference of the System Dynamics Society, 20 July - 24 July 2014, Delft, The Netherlands.
[BLOC2016] Block J.; 2016; A hybrid modeling approach for incorporating behavioral issues into workforce planning;
Proceedings of the IEEE Conference on Systems, Man, and Cybernetics (SMC) 2016, pp. 1-6.
[GUER2011] Guerry M.-A.; 2011; Hidden heterogeneity in manpower systems: A Markov-switching model approach;
European Journal of Operational Research, Vol. 210, pp. 106-113.
[JIAN2012] Jiang K., Lepak D.P., HuJ., and Baer J.C.; 2012; How Does Human Resource Management Influence Organizational
Outcomes? A Meta-Analytic Investigation of Mediating Mechanisms; Academy of Management Journal, Vol. 55, No. 6, pp.
1264-1294.
[NASI2012] Nasiripour A., Afshar Kazemi M., and IzadiA.; 2012; Effect of Different HRM Policies on Potential of employee
Productivity; Research Journal of Recent Sciences, Vol.1, No. 6, pp. 45-54.
[STER2000] StermanJ.D.;2000; Business Dynamics — Systems Thinking and Modeling for a Complex World; McGraw-Hill,
Boston, USA.
[ZEIG2000] Zeigler B. P., Praehofer H., and Kim T. G.; 2000; Theory of Modeling and Simulation - Integrating Discrete Event
and Continuous Complex Dynamic Systems; Second edition, Academic Press, San Diego, CA, USA.
