Write the Mathematical Formulation of Transportation problem.

Objective:The objective of the transportation problem is to minimize the total cost of

transporting goods from various sources (origins) to various destinations.
Parameters:
m: Number of sources.
n: Number of destinations.
sᵢ: Supply (amount of goods available) at source i, where i = 1, 2, ..., m.
dⱼ: Demand (amount of goods required) at destination j, where j = 1, 2, ..., n.
cᵢⱼ: Cost of transporting one unit of goods from source i to destination j.
Decision Variables:
xᵢⱼ: Quantity of goods transported from source i to destination j.
Mathematical Model:
Minimize:
Z = ΣᵢΣⱼ cᵢⱼxᵢⱼ
Where the double summation is taken over all i = 1, 2, ..., m and j = 1, 2, ..., n.
Subject to:
Supply Constraints:
The total quantity of goods transported from each source cannot exceed its supply:
Σⱼ xᵢⱼ ≤ sᵢ for all i = 1, 2, ..., m
Demand Constraints:
The total quantity of goods transported to each destination must satisfy its demand:
Σᵢ xᵢⱼ ≥ dⱼ for all j = 1, 2, ..., n
Non-negativity Constraints:
The quantity of goods transported must be non-negative:
xᵢⱼ ≥ 0 for all i = 1, 2, ..., m and j = 1, 2, ..., n
Balanced Transportation Problem:
A transportation problem is said to be balanced if the total supply equals the total
demand:
Σᵢ sᵢ = Σⱼ dⱼ
If the problem is unbalanced (i.e., total supply does not equal total demand), dummy
sources or destinations can be introduced to balance it.
4. Write an algorithm to solve assignment problem (by Hungerian method.)
The algorithm to solve the Assignment Problem, commonly known as the Hungarian
Algorithm:
Hungarian Algorithm for Solving the Assignment Problem
The Assignment Problem seeks to find the optimal assignment of n tasks to n resources
(e.g., workers to jobs) to minimize the total cost (or maximize the total profit/utility).I
nput: An n x n cost matrix, where each element cᵢⱼ represents the cost of assigning
resource i to task j.
Steps: 1. Check for Balanced Problem:
Ensure that the number of rows (resources) equals the number of columns (tasks). If not,
add dummy rows or columns with zero costs to balance the problem.
2. Row Reduction:
For each row, find the smallest element.
Subtract this smallest element from all elements in that row.
This step ensures that each row has at least one zero.
3. Column Reduction:
For each column, find the smallest element.
Subtract this smallest element from all elements in that column.
This step ensures that each column has at least one zero.
4. Covering Zeros: Draw the minimum number of horizontal and vertical lines to cover all
the zeros in the reduced cost matrix.
5. Optimality Test: Let k be the number of lines drawn in the previous step.
If k = n, an optimal assignment is possible, and go to step 7.
If k < n, an optimal assignment is not yet possible, and go to step 6.
6. Improve the Matrix (Iteration):
Find the smallest uncovered element (i.e., the smallest element not covered by any line).
Let this value be θ.
Subtract θ from all uncovered elements.
Add θ to the elements at the intersection of two lines (i.e., elements covered by both a
horizontal and a vertical line).
Elements covered by only one line remain unchanged.
Go back to step 4 (Covering Zeros).
7. Make the Assignment: Examine the final reduced cost matrix.
Select a zero element cᵢⱼ such that:
Row i and column j do not have any other assigned zero.
Assign resource i to task j.
Cross out (or mark as unavailable) all other zeros in row i and column j.
Repeat this process until all resources are assigned to tasks.If it's not possible to find a
unique zero in a row or column, it indicates that there might be multiple optimal solutions.
You can choose any of the available zeros.
8. Calculate the Minimum Cost:
Using the original cost matrix, sum the costs of the assignments made in step 7. This
sum is the minimum total cost of the optimal assignment.
Example: of assignment problem (by Hungerian method.)
Let's say you have the following cost matrix:
A B C D
1 9 2 7 8
2 6 4 3 7
3 5 8 1 8
4 7 6 9 4
Following the Hungarian Algorithm:
1. Balanced: The problem is balanced (4x4 matrix).
2. Row Reduction:
7 0 5 6
3 1 0 4
4 7 0 7
3 2 5 0
3. Column Reduction: (No change in this case as each column already has a 0)
4. Covering Zeros: (You'd draw lines to cover all zeros - the minimum number)
5. Optimality Test: (Check if the number of lines equals 4. If not, go to step 6)
6. Improve Matrix: (Find smallest uncovered, subtract from uncovered, add to
intersections) (Steps 4-6 are repeated until you can cover all zeros with 4 lines)
7. Make Assignment: (Select zeros such that each row/column has only one assignment)
8. Calculate Cost: (Sum the original costs for the assigned pairs)
3. Give any two fields where operations research is used.
Operations Research (OR) is a versatile discipline with applications across numerous
fields. Here are two prominent examples:
Supply Chain Management and Logistics: OR techniques are extensively used to
optimize various aspects of the supply chain. This includes:
Inventory Management: Determining optimal inventory levels to minimize holding costs
and avoid stockouts.
Transportation and Distribution: Finding the most efficient routes for delivery vehicles,
optimizing warehouse locations, and managing logistics networks to reduce
transportation costs and improve delivery times.
Production Planning and Scheduling: Deciding on production quantities, sequencing
jobs on machines, and scheduling resources to maximize efficiency and meet demand.
Finance and Investment: OR provides powerful tools for decision-making in the
financial sector:
Portfolio Optimization: Selecting a mix of assets that maximizes returns for a given level
of risk (or minimizes risk for a given return target).
Risk Management: Developing models to assess and mitigate financial risks.
Financial Forecasting: Building models to predict future financial trends and aid in
investment decisions.
Algorithmic Trading: Designing automated trading strategies based on mathematical
models and optimization techniques.

Write two applications of Linear programming problem

Product Mix Optimization (Resource Allocation):
Problem: A company produces several different products using limited resources such
as raw materials, labor, and machine time. The goal is to determine the optimal quantity
of each product to produce in order to maximize profit, given the constraints on available
resources and production capacities.
LP Formulation: The objective function would represent the total profit (sum of profit per
unit of each product multiplied by the quantity produced). The constraints would
represent the limitations on each resource (e.g., total raw material used cannot exceed
the available raw material). The decision variables would be the quantities of each
product to be produced.
Diet Planning (Cost Minimization):
Problem: A nutritionist or a food manufacturer wants to determine the most cost-effective
combination of different food items that satisfies certain nutritional requirements (e.g.,
minimum levels of vitamins, minerals, and calories).
LP Formulation: The objective function would represent the total cost of the diet (sum of
cost per unit of each food item multiplied by the quantity of that item). The constraints
would represent the minimum and/or maximum levels of each required nutrient. The
decision variables would be the quantities of each food item to include in the diet.
How an assignment problem with certain restrictions can be solved?
Solving an Assignment Problem with certain restrictions in Operations Research
requires modifications to the standard Hungarian Algorithm or the formulation of the
problem. The approach depends on the nature of the restrictions. Here are some
common types of restrictions and how to handle them:
Forbidden Assignments: Assign a Very High Cost: In the cost matrix, assign a very
large cost (practically infinity, or a sufficiently large number) to the cell corresponding to
the forbidden assignment. When the algorithm tries to minimize the cost, it will effectively
avoid these assignments in the optimal solution.
Mandatory Assignments: Method 1: Reduce the Problem Size: If resource i must be
assigned to task j, make this assignment ,reduce the problem size by removing row i and
column j. Solve the remaining (n-1) x (n-1) assignment problem. The total cost will be the
cost of the mandatory assignment plus the optimal cost of the reduced problem.
Capacity or Requirement Restrictions: Reformulation as Transportation Problem: If
these restrictions are significant, it's often more appropriate to model the problem as a
Transportation Problem. Resources become "sources" with capacities, and tasks
become "destinations" with requirements. The assignment costs become the
transportation costs.
Order or Precedence Restrictions: These types of restrictions make the problem more
complex and might not be directly solvable by the standard Assignment Algorithm. You
might need to explore: Integer Programming Formulations: Model the problem as an
integer linear program with additional constraints to represent the order or precedence
requirements.
Specialized Algorithms: For specific types of order restrictions, there might be
specialized algorithms or heuristics.
Define degeneracy in transportation problem In operations research
In Operations Research, degeneracy in a transportation problem occurs when the
number of allocated cells (basic variables) in a feasible solution is less than m + n - 1,
where:
m is the number of sources (rows in the transportation table).
n is the number of destinations (columns in the transportation table).
Why is m + n - 1 important?
In a transportation problem with m sources and n destinations, there are m + n variables
(allocations x ij ). However, due to the supply and demand constraints, one of the
equations is linearly dependent on the others. Therefore, a basic feasible solution (BFS)
should ideally have m + n - 1 independent non-negative allocations. These allocations
form the basis for moving towards an optimal solution using methods like the Stepping
Stone or MODI (Modified Distribution) method.

Mathematical Formulation of the Assignment Problem i=1 bottom , n = top

Objective: The objective of the assignment problem is to assign n tasks to n resources (e.g., workers,
machines) in such a way that the total cost (or time, or any other relevant measure) is minimized (or
maximized).
Parameters:
n: Number of tasks (and also the number of resources, as it's a square problem).
cᵢⱼ: Cost (or value) of assigning resource i to task *j`, where i,j∈ {1,2,...,n}.
Decision Variables:
xij: A binary variable such that:
xij=1 if resource i is assigned to task *j$.
xij=0 if resource i is not assigned to task *j$.
Mathematical Model (Minimization):
Minimize:
Z=i=1∑nj=1∑ncijxij
Subject to the following constraints:
Each resource is assigned to exactly one task: j=1∑nxij=1for all i=1,2,...,n
Each task is assigned to exactly one resource: i=1∑nxij=1for all j=1,2,...,n
Binary constraint on decision variables: xij∈ {0,1}for all i,j=1,2,...,n
Mathematical Model (Maximization):
If the objective is to maximize a value (e.g., profit, efficiency), the formulation changes to:
Maximize:
Z=i=1∑nj=1∑nvijxij
where vij is the value of assigning resource i to task *j$.