Upload
andreas-groessler
View
417
Download
0
Embed Size (px)
DESCRIPTION
A presentation that talks about using system dynamics in an HR management context
Citation preview
Supporting Long-term Workforce Planning
with a Dynamic Aging Chain Model:
A Case Study from the Service Industry
with two more recent additions
Andreas Größler
The case company: air traffic control
© spiegel-online.de
Long-term personnel capacity planning
as a crucial success factor
Quote Eurocontrol:
• “Manpower Planning (MP) is
necessary to meet the strategic
objective:
‘the provision of the right number of staff, with the right
qualification, at the right time and in the right place to meet
business requirements’”.
Goals of modelling and simulation
project
• Conduct a structural analysis of the existing long-
term personnel planning process for air traffic
controllers;
• provide a dynamic analysis of the existing planning
policies;
• construct a scenario-tool to improve the existing
planning policies as well as the established risk
management approach accompanying the existing
processes.
A simple system dynamics model of
capacity planning
#ATCOs at
training
#ATCOs
operative recruiting graduating leaving
recruitment time training time productive time
AIR TRAFFIC requirements
#ATCOs
needed
Some results from simple model
#ATCOs operative
6,000
4,500
3,000
1,500
0
2006 2010 2014 2018 2022 2026 2030 2034
Time (Year)
AT
CO
#ATCOs needed
600
450
300
150
0
2006 2010 2014 2018 2022 2026 2030 2034
Time (Year)
AT
CO
Independent from the scenario used, there are variations in some key
variables over time that are not easy to understand.
AIR TRAFFIC requirements)
6,000
4,500
3,000
1,500
0
2006 2010 2014 2018 2022 2026 2030 2034
Time (Year)
AT
CO
No Cycle With Cycle
Time lag through training process
Start of
process
Signing of
contract
Start of
training
Varies for
each trainee
(Ø 24 Month)
Identical for
all trainees
(15 Month) Varies for
each trainee
(Ø 6 Month)
End of
training, start
of OJT
End of OJT
Varies for
each trainee
(Ø 6 Month)
t = 0 t = 6 t = 12 t = 27 t = 51
Resulting delay behaviour: average is
longer than they think
Ordered ATCO
Signed ATCO
ATCO in IT
ATCO in OJT
ATCO
OJT 18 Months
0 10 20 30 40 50 60 70 80 90 100
Time (Month)
100
75
50
25
0
pe
rso
n
100
75
50
25
0
0 10 20 30 40 50 60 70 80 90 100
Time (Month)
pe
rson
OJT 24 Months
t≈68 t≈58
Results from client‘s perspective
• A more detailed planning paradigm can be
implemented (group level instead of centre level);
• the personnel planning cycle can be repeated several
times a year instead of only going through the process
once a year;
• the risk management can be complemented by some
quantitative scenarios that are provided almost in real-
time;
• intensified communication between all stakeholders;
• the new scenario tool can act as a learning platform for
the case company as it integrates the experience and
perspective of several departments.
Addition 1: Chains of entities…
A general issue resulting from the case
There is some confusion about the different types of supply
lines (quotes from HRM review process):
“The firm is a logistics service provider and is a service firm. The
issues at this firm are similar to issues faced by a service firm.
Service firms, similar to a logistics provider do not have a physical
product and definition of inventory is very different.”
“Supply chain in service firms are different and have their own
specific issues. References relate to service chain issues in a
service firm…”
Structural similarity of the three types
of chains
Material Work in
progress
Finished
goods purchasing fabricating assembling shipping
Proposal Draft Final
report accepting drafting finalizing delivering
Newly
hired
In
training
Fully
productive hiring starting
training
finalizing
training leaving
Physical goods supply line
Service supply line
Personnel supply line
Be aware of the ethical issue qualitative vs. quantitative individualism (Simmel)
Prototypical behaviour of chains
20
15
10
5
0 3 3 3 3 2
2
2 2
2
2 2 3 3 3 3 2 2 2 2 2 2 2 3
3
3
3 3
3
3 3 1 1
1
1 1
1
1 1 1 1 1 1 1 1 1 0 5 10 15 20 25 30 35 40 45 50
time steps
en
titie
s
1st stock 1 1 1 1 1 1 1 1 1 1
2nd stock 2 2 2 2 2 2 2 2 2
3rd stock 3 3 3 3 3 3 3 3 3
Perceived differences and structural
similarity
• Perceived differences of the three types of chains, in
particular regarding
– Utilization of “production” capacities
– Premature outflow from the chain
– Divisibility of entities
• Because of structural similarity, differences are mainly
caused by
– Inappropriate mix of supply line elements with attributes of these
elements (“co-flow”)
– The three types of supply lines regularly are located at different
organisational levels
• “Strategic architecture” (Warren, 2007)
Addition 2: Forms of delays…
Female professors task
• The analytics of a gender quota
• Dutch university: balance number of male and female
professors
• Participants have shown gross mis-estimations
(Bleijenbergh et al. 2011)
• Influence of political loadedness of task?
Discrete vs. continuous delays
= two experimental groups
x x
to t* t to t*
t
input system response
Task structure in system dynamics
notation
Male
professors
Female
professors
hiring male profs
hiring female profs
leaving male profs
leaving female profs
avg time at
university percentage female profs
hirings necessary
Ini Male
Ini Female
Estimations do not differ between
experimental groups
32%
41%
27% outside bounds (% < 50 or > 100)
wrong estimate (diff. > 5 years)
correct estimate (diff. <= 5 years)
No statistical differences between experimental groups for values
of estimations participants do not differentiate between discrete
and continuous delays.
Average error between experimental
groups differs a lot, though
0
1
2
3
4
5
6
7
8
9
discrete continuous
years
Significant statistical differences between groups for goodness of estimations (estimations compared to “true” solutions derived from respective simulation model – discrete
vs. continuous delay version).