OPERATION RESEARCH

L-2-3-4-Formulation of LP
Contents
• LP Formulations
• Scheduling problems
• Transportation Problems
• Assignment Problems
• Multi-period Problems
• Balancing Problems
• Assumptions of LP
 Scheduling Problems
• Scheduling problems are common in the field of Operations Research
• Used to determine the most efficient way to allocate resources and time to various tasks.
• There are different types of scheduling problems, and the solution methods can vary
depending on the specific characteristics of the problem.
• Common types of scheduling problems:
1. Job Scheduling:
• Single Machine Scheduling: One machine processes a set of jobs. The objective is
usually to minimize makespan (total completion time).
• Parallel Machine Scheduling: Multiple machines process a set of jobs
simultaneously. The goal is often to minimize makespan or the total time all machines spend
processing jobs.
2. Project Scheduling:
• Project Network Analysis (PERT/CPM): Involves scheduling and coordinating the
activities of a project.
It helps to identify the critical path and estimate project duration.
 LP formulation for Scheduling Problems

• Linear Programming (LP) formulation for scheduling problems involves

• Allocating resources (like workers, machines, or time slots)
• to activities (like jobs, tasks, or shifts)
• in a way that optimizes a certain objective (like minimizing costs, maximizing
efficiency, or meeting deadlines).
 Example: Machine Scheduling Problem
• Suppose a factory have two machines, Machine A and Machine B, and three jobs, Job 1, Job
2, and Job 3. Each job can be processed on either machine, but each machine can only handle
one job at a time. Your goal is to minimize the total completion time (makespan) for all jobs.
• Problem Information:
• Processing Times:

Job 1 Job 2 Job 3

Machine A 4 hr 2 hr 6 hr
Machine B 5 hr 3 hr 4 hr
Example: Machine Scheduling Problem
• Decision Variables
• Let
𝑥𝑖𝑗 be a binary variable
where: 𝑥𝑖𝑗=1 if Job j is assigned to Machine 𝑖 , 𝑥𝑖𝑗= 0 otherwise
• For this example,
𝑥𝐴1 =1 [ if Job 1 is assigned to Machine A , 0 otherwise]
𝑥𝐴2 =1 [ if Job 2 is assigned to Machine A , 0 otherwise]
𝑥𝐴3 =1 [ if Job 1 is assigned to Machine A , 0 otherwise]
𝑥B1 =1 [ if Job 1 is assigned to Machine B , 0 otherwise]
𝑥B2 =1 [ if Job 2 is assigned to Machine B , 0 otherwise]
𝑥B3 =1 [ if Job 1 is assigned to Machine B , 0 otherwise]
 Example: Machine Scheduling Problem
• Constraints
• Each job is assigned to exactly one machine:
xA1 + xB1 = 1
xA2 + xB2 = 1
xA3 + xB3 = 1
• Each machine can process only one job at a time:

xA1+xA2+xA3 ≤ 1
xB1+ xB2+xB3 ≤ 1
•
 Example: Machine Scheduling Problem
• Objective Function
• Minimize the makespan. The makespan is the maximum completion time of the jobs on both
machines:
Makespan = max (Completion Time on Machine A, Completion Time on Machine B)

Let 𝐶𝐴 and 𝐶𝐵 represent the total processing times on Machine A and Machine B, respectively.
The objective function becomes :
Minimize max(CA, 𝐶𝐵)
Where

CA = 4xA1+ 2xA2+ 6xA3

CB = 5xB1+ 3xB2+ 4xB3
 Example: Machine Scheduling Problem
• Formulation Summary
• Objective Function:
Minimize Max(4xA1+ 2xA2+ 6xA3 + 5xB1+ 3xB2+ 4xB3 )

Subject to:
xA1 + xB1 = 1
xA2 + xB2 = 1
xA3 + xB3 = 1
xA1+xA2+xA3 ≤ 1

xB1+xB2+xB3 ≤ 1
𝑥𝑖𝑗∈{0,1} for all 𝑖 and 𝑗
 Transportation Problems
• A transportation problem involves finding
• the most cost-effective way to transport goods from several suppliers (sources) to
multiple consumers (destinations),
• considering the supply at each source,
• the demand at each destination,
• and the transportation cost per unit between each source and destination.
• The objective is to minimize the total transportation cost while satisfying supply and
demand constraints.
 Transportation Problems
• Here are the basic components of the Transportation Problem:
1. Suppliers (Sources):
Locations where goods are available or produced.
2. Destinations (Sinks):
Locations where goods need to be delivered.
3. Supply:
The amount of goods available at each supplier.
4. Demand:
The amount of goods required at each destination.
5. Transportation Cost:
The cost associated with transporting one unit of the product from
a supplier to a destination.
 LP formulation for Transportation Problems

• The goal is to determine the optimal transportation plan that minimizes the total
transportation cost, considering the
• supply and demand constraints. The decision variables represent the amount of goods to be
transported from each supplier to each destination.
• The transportation problem can be solved using various algorithms, such as the Simplex
Method or specialized algorithms like the Northwest Corner Method, Vogel’s
Approximation Method (VAM), or MODI Method.
 LP formulation for Transportation Problems
• Example
Suppliers (Sources):
Source 1: 20 Units available
Source 2: 30 Units available
Consumers (Destinations):
Destination A: 25 Units needed
Destination B: 15 Units needed
Destination C: 10 Units needed
Transportation Costs (per unit):

Destination A Destination B Destination C

Source 1 4 3 6
Source 2 5 2 7
Example: Transportation Problem
• Decision Variables
• Let
𝑥𝑖𝑗 be the number of units transported from Source i to destination j
Where
𝑥1A = number of units transported from Source 1 to Destination A
𝑥1B = number of units transported from Source 1 to Destination B
𝑥1B = number of units transported from Source 1 to Destination C
𝑥2A = number of units transported from Source 2 to Destination A
𝑥2B = number of units transported from Source 2 to Destination B
𝑥2C = number of units transported from Source 2 to Destination C
 Example: Transportation Problem
• Objective Function
Minimize the total transportation cost:
Minimize Z = 4x1A+ 3x1B+ 6x1C + 5x2A + 2x2B + 7x2C
• Constraints
• Supply Constraints: (The total amount transported from each source cannot exceed its supply.)

x1A + x1B + x1C ≤ 20 [Supply from Source 1]

x2A + x2B + x2C ≤ 30 [Supply from Source 2]

• Demand Constraints: (The total amount transported to each destination must meet its demand.)
x1A+ x2A = 25 [Demand at Destination A]
x1B+ x2B = 15 [Demand at Destination B]
x1C+ x2C = 10 [Demand at Destination C]
• Non-negativity Constraints:
xij ≥ 0 for all i , j
 Example: Transportation Problem
• LP formulation Summary
• Objective Function:

Minimize Z = 4x1A+ 3x1B+ 6x1C + 5x2A + 2x2B + 7x2C

• Subject to

x1A + x1B + x1C ≤ 20 [Supply Constraints from Source 1]

x2A + x2B + x2C ≤ 30 [Supply Constraints from Source 2]

x1A+ x2A = 25 [Demand Constraint at Destination A]

x1B+ x2B = 15 [Demand Constraint at Destination B]

x1C+ x2C = 10 [Demand constraint at Destination C]

xij ≥ 0 [Non- Negativity Constraints]

 Assignment Problems
• It deals with the assignment of a set of tasks (or jobs) to a set of agents (or resources) in a
way that minimizes the total cost or time required for completion. The problem assumes that
each task must be assigned to exactly one agent, and each agent can handle at most one
task.
• key components of the Assignment Problem:
1. Agents (Resources):
Individuals or entities capable of performing tasks.
2. Tasks (Jobs):
Activities or assignments that need to be completed.
3. Cost (or Time):
The measure associated with assigning a specific task to a particular agent.
The goal is to find the optimal assignment that minimizes the total cost or time.
 LP formulation for Assignment Problems

• Example
Suppose there are 3 workers (W1, W2, W3) and 3 tasks (T1, T2, T3). The cost matrix for
assigning workers to tasks is given as follows:

T1 T2 T3
W1 10 12 18
W2 14 15 13
W3 16 11 17
 Example: Assignment Problem
• Decision Variables
• Let
𝑥𝑖𝑗 be a binary variable
where: 𝑥𝑖𝑗=1 if worker i is assigned to task j , 𝑥𝑖𝑗= 0 otherwise
• For this example,
𝑥11 =1 [ if worker 1 is assigned to Task 1 , 0 otherwise]
𝑥12 =1 [ if worker 1 is assigned to Task 2, 0 otherwise]
𝑥13 =1 [ if worker 1 is assigned to Task 3, 0 otherwise]
𝑥21 =1 [if worker 2 is assigned to Task 1, 0 otherwise]
𝑥22 =1 [if worker 2 is assigned to Task 2, 0 otherwise]
𝑥23 =1 [if worker 2 is assigned to Task 3, 0 otherwise]
𝑥31 =1 [if worker 3 is assigned to Task 1, 0 otherwise]
𝑥32 =1 [if worker 3 is assigned to Task 2, 0 otherwise]
𝑥33 =1 [if worker 3 is assigned to Task 3, 0 otherwise]
 Example: Assignment Problem
• Objective Function
Minimize the total assignment cost
Minimize 𝑍= 10𝑥11+12𝑥12+18𝑥13+14𝑥21+15𝑥22+13𝑥23+16𝑥31+11𝑥32+17𝑥33
• Constraints
• Each worker is assigned exactly one task:
x11 + x12 + x13 = 1 [Worker 1]
x21 + x22 + x23 = 1 [Worker 2]
x31 + x32 + x33 = 1 [Worker 3]
• Each task is assigned exactly one worker:
x11 + x21 + x31 = 1 [Task 1]
x12 + x22 + x32 = 1 [Task 2]
x13 + x23 + x33 = 1 [Task 3]
• Binary Constraints:
𝑥𝑖𝑗 ∈ {0,1} for all i , j
 Multi-period Problems
• A Multi-Period Problem generally refers to an optimization or decision problem where decisions
need to be made over multiple time periods or stages. This concept is often applied in Operations
Research, dynamic programming, and various areas of management and engineering. A few
examples to illustrate the idea of multi-period
1. Multi-Period Production Planning:
Decision makers need to determine the production levels for each product in each
period to meet demand while considering factors like inventory holding costs, production
setup costs, and production capacity.
2. Multi-Period Financial Planning:
In finance, individuals or companies may face decisions over multiple time periods, such as
investment planning, portfolio management, or budget allocation over several years.
3. Multi-Period Inventory Management:
Businesses with inventory may need to make decisions on how much to order, when to order,
and how much to keep in stock over several periods, taking into account carrying costs, order
costs, and demand variability.
 LP formulation for Multi-period Problem
• Example
Consider a company that needs to plan its production over three months to meet the demand
for its product while minimizing costs. The company can produce units in each month, but
there are costs associated with production, storage, and unmet demand. The following data is
provided by the company to minimize the total cost

June July August

Demand of production 100 units 150 units 130 units
Production Cost 5 Rs/Unit 6 Rs /Unit 7 Rs/Unit
Storage Cost 1 Rs 1 Rs 1 Rs

The maximum production capacity in each month is 160 Units and initial inventory is 20 units
 Example: Multi-period Problem
• Decision Variables
• Let
𝑥t = Number of units produced in month 𝑡 (where 𝑡 = 1, 2, 3 for June, July, August,
respectively)
It = Inventory level at the end of month t
• Objective Function
Minimize the total cost, which includes production and storage costs over the three
months:

Minimize 𝑍= 5𝑥1 +6𝑥2+7𝑥3+ 1I1+ 1I2+ 1I3

 Example: Multi-Period Problem
• Constraints
• Inventory Balance Constraints:
The inventory at the end of each month is determined by the previous month's
inventory, the production in the current month, and the demand in the current month:
20 + x1 - 100 = I1 [Inventory at the end of June]
I1 + x2 - 150 = I2 [Inventory at the end of July]
I2 + x3 - 100 = I3 [Inventory at the end of August]
• Production Capacity Constraints:
The number of units produced each month cannot exceed the production capacity:
x1 ≤ 160 , x2 ≤ 160 , x3 ≤ 160
• Non- Negativity Constraints:
𝑥t ≥ 0 , I t ≥ 0 for t = 1,2,3
 Balancing Problems
• In Operations Research, balancing problems typically refer to problems where the goal is to
allocate resources or tasks in a way that achieves a balance across various criteria or constraints.
The term can be applied to different contexts, and here are a few examples
• Balanced Transportation Problem:
In logistics and supply chain management, balancing the transportation of goods from
suppliers to destinations is essential. The objective is to allocate resources in such a way that
minimizes costs while meeting supply and demand requirements.
• Resource Balancing in Project Management:
In project management, balancing resources such as manpower, equipment, and materials is
crucial for meeting project deadlines and minimizing costs. This involves allocating resources
efficiently across different project tasks.
3. Balanced Assignment Problem:
This is a variant of the assignment problem where the goal is not only to minimize costs but
also to balance the workloads among the agents. This could be relevant in tasks like employee task
assignments.
LP formulation for Balancing Workforce Scheduling Problem

• Example
Suppose a company needs to schedule workers across 5 days (Monday to Friday) to cover
required daily shifts while minimizing the cost of assigning employees and ensuring that the
workload is evenly distributed. The daily shifts requirements are on Monday: 3 shifts ,
Tuesday: 2 shifts , Wednesday: 4 shifts , Thursday: 2 shifts and Friday: 3 shifts . Further more
the cost of per shift of each worker is given below

Worker 1 Worker 2 Worker 3

Cost per shift 100 Rs/Unit 120 Rs /Unit 110 Rs/Unit
 Example: Balancing Problem
• Decision Variables
• Let
𝑥ij represents the number of shifts assigned to the worker i on day j
Where
𝑥1 1 = shifts assigned to Worker 1 on Monday
𝑥1 2 = shifts assigned to Worker 1 on Tuesday
𝑥ij for all combinations of workers i and days j.
• Objective Function
Minimize the total cost of assigning workers to shifts:
Minimize 𝑍= 100 (𝑥11 + 𝑥12+𝑥13 +𝑥14 + 𝑥15) + 120(𝑥21 + 𝑥22+𝑥23 +𝑥24 + 𝑥25) + 110(𝑥31 + 𝑥32+𝑥33 +𝑥34 + 𝑥35)
 Example: Balancing Problem
• Constraints
• Shift Requirements Constraints:
Ensure the required number of shifts is covered each day:
x11 + x21 + x31 = 3 [ Monday]
x12 + x22 + x32 = 2 [Tuesday]
x13 + x23 + x33 = 4 [Wednesday]
x14 + x24 + x34 = 2 [Thursday]
x15 + x25 + x35 = 3 [Friday]
• Balancing Constraints:
Ensure that the workload is balanced among workers:
x11 + x12 + x13 + x14 + x15 ≤ M [Worker 1]
x21 + x22 + x23 + x24 + x25 ≤ M [Worker 2]
x31 + x32 + x33 + x34 + x35 ≤ M [Worker 3]
Where M is representing the Maximum shifts any worker is assigned. The objective is to minimize M to
achieve balance.

• Non- Negativity Constraints:

𝑥ij ≥ 0 for all i , j
 Assumptions Of Linear Programming
• Linear Programming (LP) is a mathematical optimization technique that is widely used for
decision-making in various fields. When formulating and solving linear programming problems,
certain assumptions are typically made. Here are the key assumptions of Linear Programming:
1. Linearity:
The relationships between decision variables and the objective function, as well as the
constraints, are assumed to be linear. This means that each decision variable appears to the first
power and is multiplied by a constant coefficient.
2. Additivity:
The total contribution of each decision variable to the objective function and constraints is the
sum of its individual contributions. In other words, the effect of changing one decision variable is
independent of the levels of other decision variables.
3. Divisibility:
LP assumes that decision variables can take any non-negative real value, including fractional
values. This assumption is not always realistic in practical situations where decision variables
might represent discrete entities (e.g., the number of machines or units produced).
 Assumptions Of Linear Programming
4. Certainty:
Linear programming assumes that all parameters, such as coefficients in the objective function
and the right-hand sides of constraints, are known with certainty and do not change during the
decision-making process.
5. Non-Negativity:
Decision variables are assumed to take non-negative values. This is a fundamental assumption,
as negative quantities often have no practical interpretation in many real-world problems..
6. Independence:
The decision variables are assumed to be independent, meaning that changes in the value of
one variable do not affect the coefficients of other variables in the objective function or
constraints.
7. Proportionality:
The coefficients in the objective function and constraints are assumed to be constant and do not
vary proportionally with the scale of the problem.
 Assumptions Of Linear Programming
8. Rationality:
Linear programming assumes that decision-makers are rational and will make choices that
maximize (or minimize) the objective function based on the available information and the given
constraints.
9. Deterministic Environment:
LP assumes a deterministic environment, where all input parameters are known with certainty
and are not subject to variability or randomness. This assumption might be relaxed in stochastic
programming for problems involving uncertainty.