Chapter 19 Business Dynamics by John D. Sterman

1
Chapter 19Business Dynamics by John D. Sterman

• Labor Supply Chains and the Origin of Business
  Cycles

2
Chapter outline

• The Labor Supply Chain
• Interactions of Labor and Inventory Management
• Inventory-Workforce Interactions and the Business
  Cycle
• Summary

3
Labor supply chain (LSC)

• The human resource supply chain
• Omission of labor and capital in production
  model- not realistic
• We now apply the stock management structure to
  represent labor
• Hiring rate- delay
• Quit rate (Attrition rate)-delay

4
Structure of Labor and Hiring

• Referring to figure 19-1 on page 758
• LaborINTEGRAL (Hiring rate- Attrition rate,
  Laborto) ..19-1
• Attrition Rate Labor/Avg Duration of
  Employment ..19-2
• Hiring RateVacancies/Time to fill
  vacancies ..19-3
• VacanciesINTEGRAL (Vacancy Creation Rate-Vacancy
  Closure Rate, vacanciesto).. 19-4
• Vacancy Closure Rate Hiring Rate ..19-5
• Why isnt there a direct flow from vacancies to
  labor?
• What can you say about the source of hiring?

5
(No Transcript)
6
More equations of key rates

• Vacancy Creation Rate MAX(0, Desired Vacancy
  Creation rate) ..19-6
• Desired Vacancy Creation Rate Desired Hiring
  Rate adjustment for Vacancies ..19-7
• Adjustment for Vacancies (Desired
  Vacancies-Vacancies)/Time to Adjust
  Vacancies .. 19-8
• Desired Vacancies MAX(0, Exp Time to fill
  vacanciesDesired Hiring Rate) .. 19-9
• Exp time to fill Vacancies average time to fill
  Vacancies .. 19-10
• Desired Hiring rate Expected Attrition rate
  Adjustment for Labor ..19-11
• Expected Attrition rateAttrition rate ..19-12
• Adjustment for Labor(Desired Labor-Labor)/Time
  to Adjust Labor ..19-13

7
Behavior of LSC (Labor Supply chain)

• Desired labor is exogenous
• Hire time depends on several factors
• Model parameters
• Avg employment time 100 weeks(2 yrs)
• Time to fill vacancies 8 weeks
• Initial1000 workers Quit rate 10/week
• 80 vacancies (in supply line)
• Vacancies adj time 4 weeks
• Labor adj time 13 weeks
• Test run with 50 step at week 5

8
Behavior of LSC contd

• Behavior is similar to stock management structure
  on simulation
• 50 step sends desired vacancy creation rate from
  10/week to 125/week
• Think about need to down size?
• We need lay off rate and vacancy cancellation
  rate
• Both these processes have time delays
• Vacancy Cancellation rateMIN (Desired vacancy
  cancellation rate, Maximum vacancy cancellation
  rate) ..19-14
• Maximum Vacancy cancellation rate
  vacancies/Vacancy cancellation time ..19-15
• Desired Vacancy Cancellation rate MAX(0,
  -Desired Vacancy Creation Rate) ..19-16
• Similar logic is applied for formulation of
  19-17, 19-18 and 19-19 (refer page 764) in
  respect of Layoff Rate
• Willingness to lay off0 no layoff policy 1
  layoff ok

9
Interactions of Labor and Inventory Management

• Extending the model to include production
  structure from Chapter 18
• Give a 20 step increase in Customer orders
• Production starts adjust to desired starts with
  delay- labor hiring etc- so Inventory takes
  longer to adjust
• System oscillates vigorously
• Production Start rate Labor workweek Labor
  productivity ..19-20
• Desired Labor Desired Production
  Starts/(Standard workweekExpected
  Productivity) ..19-21

10
Source of Oscillation

• Refer the phase plot on page 772
• Orders jump ?production same? inventory falls.
  Firms demand goes up? desired production rises
  ?but now labor needs to rise before WIP can rise
• With delays labor finally rises? desired start
  rate customer orders. However, Inventory keeps
  falling due to delay in Production process
• Corrective forces acting on the system to adjust
  it to the shock produce the oscillation, before
  the system finally reaches new equilibrium at 20
  above the initial start point (10,000
  widgets/week)

11
Adding Over time

• Work week Standard Work week Effect of
  Schedule Pressure on workweek
• Effect of Schedule pressure on work week
  ƒ(Schedule Pressure)
• Schedule Pressure Desired Production
  Starts/Standard Production Starts
• Standard Production Starts labor Standard
  WorkweekExpected Productivity
• Schedule Pressurelt1 gt Surplus labor
• Schedule Pressuregt1 gt Shortage of labor

12
Schedule Pressure-Workweek

• Refer to the plot on page 775
• Reference line workweek1
• 45 line workweek adjusted per schedule pressure
  (overtime policy that adjusts to work pressure)
• Workweek typically lies between the two reference
  lines- not reasonable to adjust workweek to
  expect workweek to be 80 hours under pressure- or
  to let workweek fall below 30 hours under no
  pressure
• Management would like to continue production for
  stocking purpose when pressure lt1 (to keep
  workers busy, skills intact etc)

13
Flexible workweek

• Refer the new phase plot on page 778
• Flexible workweek makes it possible to reach new
  equilibrium with less oscillation than under
  fixed work week scenario
• System recovers from shock without generating
  cycles seen under constant workweek case
• Obviously the workweek can not increase
  indefinitely and stretching the work week for
  sustained periods of time may lead to undesirable
  results- Since labor is simultaneously adjusting
  to the desired level defined by current demand-
  Schedule Pressure will ease

14
Cost of instability

• There is no account by name cost of instability
• However, its well known that instability and
  oscillation are costly
• To properly assess the impact of policies one
  needs to identify costs of instability and
  include them in the model
• Conversely, define the profit function in short
  and long run and try to optimize such profit
  function variable in the model

15
More improvements/variations

• Training and Experience
• Separation of rookies and experts
• policy of lay offs- quit rates etc
• On the job training delays and costs
• Think of industries where trained staff are not
  readily available/training time is too long
  (aerospace, machine tools)

16
Business Cycles

• Business Cycles are caused by the fundamental
  structure of the industrial economy rather than
  other factors
• IT has helped in several ways to reduce the
  instability by providing instantaneous
  information (delay reduction)
• Lean manufacturing, JIT etc also help reduce the
  instability