8.1EM (Lec11-13) Quantitative DM-LP Optimization
8.1EM (Lec11-13) Quantitative DM-LP Optimization
• Engineering managers, middle and senior tiers, use a number a qualitative and
quantitative tools in their planning and decision making process.
• Some important tools for strategic planning are:
– Market research
– SWOT analysis, PEST/PESTLE, Porters 5-forces analysis
– Decision tree
– Risk assessment (Monte Carlo simulation, what-if analysis )
– Failure Mode & Effect Analysis (FMEA), Root Cause Analysis (RCA), ----
– Product life cycle analysis
– Performance benchmarks
• Customer related measures
• Process related measures
• Financial measures
• Employee related measures
• Competition related measures
• Tactical Planning tools include project management tools (CPM, PERT, etc), OR (LP,
simplex method, Integer programing, etc), forecasting, etc
• Operational planning include work study, time study, time-motion study, OR (LP), etc
DM Tools for Management
• Qualitative and Quantitative tools, as required, are used
for DM to find feasible and optimum solution to the problem
• In situations where variables / factors cannot be defined
properly quantitatively, qualitative DM tools are preferred
• Quantitative analysis uses a scientific approach to DM
process.
• Both types of DM tools can be used simultaneously for
addressing different aspects of the same problem.
• Quantitative tools include two types of models:
– A model is deterministic if we know all values with certainty.
– Models that involve chance of risk, often measured as probability
value, are called probabilistic models.
Quantitative Tools for
Management
Quantitative DM Approach
Defining the Problem Develop a clear & concise statement of the problem
Analyzing the results Workout implications of the obtained solution and check
sensitivity of the model with different input data and
mathematical model (sensitivity analysis)
Implementing results
Advantages of Mathematical Modeling
• Models can accurately represent reality.
• Models help decision maker in formulating problems and
visualizing its effects by changing the input values, by
interpolating and extrapolating.
• Models can give us insight and other useful information.
• Models help in saving time, money and other resources in
making accurate decisions and resolving problems.
• Model may be the only way to study, predict and solve large
and complex problems in a timely way without much expense.
• Model used help in communicating problems, related variables
and its solutions to other in a very clear way.
Operational Research
(Linear Programming)
• Operations Research (OR) deals with quantitative
modeling of complex operations and uses these models to
support DM in any industry or public services. It is another
name for optimization.
• Its primary aim is optimum utilization of resources in any
undertaking under a given situation so that objective
function may be maximized or minimized.
– profits/services/manufacturing resource utilization strategy, etc
may be maximized
Formulation of LP Problem
1. Maximization of profit from product mix
2. Let x1 and x2 are # of products P1 and P2
respectively
3. Define Objective function;
Maximum Profit = 50 . x1 + 40 . x2
4. Identify resources and constraints;
5 . x1 + 3 . x2 ≤ 350
3 . x1 + 3 . x2 ≤ 270
x1 ≥ 0
x2 ≥ 0
LPP – Mfg mix case study
A company ABC makes two products, P1 and P2. Each P1 requires 5 kg of
material M and 3 kg of material N. Each P2 requires 3 kg of material M and
3 kg of material N. In the warehouse, 350 kg of M and 270 kg on N are
available. The profit is $50 for each P1 and $40 for each P2. What mix of
P1 and P2 should the company manufactures in order to maximize profit.
• Solution space is bounded by an
area OABCO
• Profit = P = 50 . x1 + 40 . x2
• Calculate profit at corner points
of feasible region
Point O (0,0); P = $0
Point A (0,90); P = $3600
Point B (40,50); P = $4000
Point C (70,0); P = $3500
• Optimum profit is at Point B
LPP – Mfg mix case study
• Solution space is bounded by an
area OABCO
• Profit = P = 50 . x1 + 40 . x2
• Calculate profit at corner points
of feasible region
Point O (0,0); P = $0
Point A (0,90); P = $3600
Point B (40,50); P = $4000
Point C (70,0); P = $3500
What is Profit line or Iso Profit line??
• This method assigns a value to objective function, profit in this case, let say profit = $1000
• P = 1000 = 50.x1 + 40.x2 , and calculate values for decision variables x1 and x2
• For x1=0, x2 = 1000/40 = 25 and for x2=0, x1 = 1000/50=20. Iso profit line is drawn between
points (0,25) and (20,0)
• The process is repeated until we get to profit value which corresponds to a corner point that
yields optimum value within the feasible region
• Conceptually this method is similar to corner points method of feasible region but requires
more steps/calculations to get to the solution.
Special cases in LPP
There are four special cases when using graphical method
to solve LPP;
– Infeasibility - lack of feasible solution region can occur if
constraints conflict with one another
– Unboundedness - open ended feasible region can occur
without violating any constraints, leading to very high value of
objective function
– Redundancy - A redundant constraint is one that does not affect
the feasible solution region
– Alternate optimal solutions - multiple optimal solutions are
possible when the objective function isoprofit or isocost line runs
perfectly parallel to one of the constraint of the problem
Infeasibility Unboundedness
• Most real life problems have more than two variables; may be
dozens, hundreds, thousands-- and constraints may go as high
as hundreds, thousands, etc-- requiring numerical algorithms to
solve.
– Even in this case, the optimal solution would still lie at a corner
point of many sided, many-dimensional figure (n-dimension
polyhedron) that represents the area of the feasible solutions.
Delta airlines reported savings of 0.25Millon$ per day by optimizing its flight
scheduling, close to a saving of 100 Million$ per year. (case study of 1994)
Case Study # 2
Purchase of new airplanes by PIA
PIA owns an aging fleet of Boeing-727 jet airplanes. It is considering a major
purchase of new models of Boeing aircrafts; B‐757 and B‐767, to maximize
passenger travelling (max seat occupancy).
The decision must take into account numerous cost and capability factors,
including the following:
A. purchase of up to 17 new models of Boeing aircrafts; B‐757 and B‐767
B. airline can finance up to $400 million in purchases; each B‐757 will cost $35
million while each B‐767 will cost $22 million
C. at least one third of the planes purchased should be longer range B‐757
D. annual budget is to be no more than $8 Million; annual maintenance cost per
B‐757 is estimated to be $800,000, and $500,000 for each B‐767
E. each B‐757 can carry 125,000 passengers per year, while each B‐767 can
fly 81,000 passengers annually.
Formulate the problem as Integer Programming problem. Use LINGO to
solve. Interpret the solution and give your comments.
Case Study # 4
Purchase of new airplanes by PIA
Constraints
A. purchase of up to 17 new models of Boeing aircrafts; B‐757 and B‐767
B. airline can finance up to $400 million in purchases; each B‐757 will cost $35
million while each B‐767 will cost $22 million
C. at least one third of the planes purchased should be longer range B‐757
D. annual budget is to be no more than $8 Million; annual maintenance cost per
B‐757 is estimated to be $800,000, and $500,000 for each B‐767
Objective Function
each B‐757 can carry
125,000 passengers per
year, while each B‐767 can
fly 81,000 passengers
annually
LPP – Diet Selection case study
https://people.richland.edu/james/ictcm/2006/3dsimplex.html
Limitations to LPP
• The optimal solution may be buy 6.7 trucks or manufacture 5.3
parts. Obviously one would buy either 6 or 7 trucks or manufacture 5
or 6 parts instead of the mixed number solution. Therefore, the
optimal solution is rounded either up or down , as appropriate.
– There is another algorithm, called Integer Programming, that provide
solution integer (whole number ) solution.
• Very often, the objective function and constraints / resources are not
relate linearly with the variables.
– Non Linear programming techniques are available to deal with such
cases.
• LP does not deal with uncertainty but real life problems often deal
with uncertainty.
– Methods such as chance constrained programming or LP under
uncertainty may be used in such cases.