Excel Solver Solving Method List: Course Material: DR A R Singla
Excel Solver Solving Method List: Course Material: DR A R Singla
Solver has been greatly revamped and improved. The primary change is the presence
of the Select A Solving Method list. From this list, you must select the appropriate
solution engine for your optimization problem:
The Simplex LP engine solves linear optimization problems. A linear optimization
problem is one in which the target cell and constraints are all created by adding
terms of the (changing cell)*(constant) form.
The GRG nonlinear engine solves optimization problems in which the target cell
or some of the constraints are not linear and are computed by using common
mathematical operations such as multiplying or dividing changing cells, raising
changing cells to a power, exponential or trigonometric functions involving
changing cells, and so on. The GRG engine includes a powerful Multistart option
that enables you to solve many problems that were solved incorrectly with previous
versions of Excel.
The Evolutionary Solver engine is used when your target cell or constraints
contain non-smooth functions that reference changing cells. A non-smooth
function is one whose slope abruptly changes. For example, when x = 0, the slope
of the absolute value of x abruptly changes from -1 to 1. If your target cell or
constraints contain IF, SUMIF, COUNTIF, SUMIFS, COUNTIFS, AVERAGEIF,
AVERAGEIFS, ABS, MAX, or MIN functions that reference the changing cells, you
are using non-smooth functions, and the Evolutionary Solver engine probably has
the best shot at finding a good solution to your optimization problem.
o Do not use more labor or raw material than is available. That is, the
values in cells D14:D15 (the resources used) must be less than or equal
to the values in cells F14:F15 (the available resources).
o Do not produce more of a drug than is in demand. That is, the values in
the D2:I2 cells (pounds produced of each drug) must be less than or equal
to the demand for each drug (listed in cells D8:18)
o You can’t produce a negative amount of any drug.
o Adding these constraints ensures that when Solver tries different
combinations for the changing cell values, only combinations that satisfy
the following parameters are considered
Next, choose Simplex LP Method, because the product mix problem is a special
type of Solver problem called a linear model as the target cell is computed by
adding together the terms of the (changing cell)*(constant) form. Each constraint
satisfies the linear model requirement. This means that each constraint is
evaluated by adding together the terms of the (changing cell)*(constant) form and
comparing the sums to a constant.
Changing cells Number of employees who start work (the first of five
consecutive days) each day of the week. Each changing cell must be a
nonnegative integer.
For each day of the week, the number of employees who are working must be
greater than or equal to the number of employees required-(Number of employees
working)>=(Needed employees)
Suppose a company produces a certain drug at its Los Angeles, Atlanta, and New
York facilities. Each month, the Los Angeles plant can produce up to 10,000
pounds of the drug. Atlatna can produce up to 12,000 pounds, and New York can
produce up to 14,000 pounds. The company must ship the number of pounds
listed in cells B2:E2 each month to the four regions of the United States-East,
Midwest, South, and West-as as shown.
Total Shipping cost (the target cell) can be calculated by using SUMPRODUCT
function on two rectangles B4:E6 (named Costs) and the changing cells range
B10:E12 (named Shipped). Therefore, total shipping and production cost is
computed in cell B18 with SUMPRODUCT (costs, shipped) formula.
To express the problem’s constraints, first compute the total shipped from each
supply point. By entering the SUM(B10:E10) formula in cell F10. Copying this
formula to F11:F12 computes the total shipped from Atlanta and New York City.
Later, you’ll add a constraint (called a supply constraint) that ensures that the
amount shipped from each location does not exceed the plant’s capacity.
Next, compute the total received by each demand point by entering the
SUM(B10:B12) formula in cell B13. By copying this formula from B13 to C13:E13,
you compute the pounds of the drug received by the Midwest, South, and West
regions. Later, you’ll add a constraint (called a demand constraint) that ensures
that each region receives the amount of the drug it requires.
Minimize total shipping cost in B18 by changing cells are the number of pounds
shipped from each plant to each region of the country (B10:E12) The
F10:F12<=H10:H12 constraint (the supply constraint) ensures that the amount
sent from each plant does not exceed its capacity. The B13:E13>=B15:E15
constraint (the demand constraint) ensures that each region receives at least the
amount of the drug it needs.
This model is a linear Solver model because the target cell is created by adding
together the terms of the form (changing cell)*(constant), and both your supply and
demand constraints are created by comparing the sum of changing cells to a
constant. Because the model is linear, choose the Simplex LP engine. Clearly,
shipments must be nonnegative, so select the Make Unconstrained Variables
Non-Negative check box.
The company want to take project that contribute the greatest NPV subject to
limited resources by determining which of 20 software projects it should undertake.
Target Cell: Maximize the NPV generated by selected projects. In cell B2, use
the SUMPRODUCT(doit, NPV) formula to compute the total NPV generated by
selected projects. (The NPV range name refers to the C6:C25 range.)
Changing cells: Look for a 0 or 1 binary changing cell for each project. Locate
these cells in the A6:A25 range (named the doit range). Implement Binary
constraint on these cells.
Constraints You need to ensure that for each Year t (t = 1, 2, 3), Year t capital
used is less than or equal to Year t capital available, and Year t labor used is less
than or equal to Year t labor available. Compute the capital used each year and
the labor used each year by copying the SUMPRODUCT(doit, E6:E25) formula
from E2 to F2:J2. The E2:J2<=E4:J4 constraint ensures that during each year, the
capital and labor used are less than or equal to the capital and labor available.
Constraints Implement constraint so that you can do only 4 projects from among
Project 1 through 10.
Solve using Simplex LP method as target cell is computed as the sum of terms
that have the (changing cell) * (constant) form
Can I use Solver to verify the accuracy of the Excel PMT function or to determine
mortgage payments for a variable interest rate?
Use FInmathsolver.xlsx
Use solver to calculate your monthly payment on a 10 month loan for $10000 at
an annual interest rate of 6.67%.
The Solver target cell is to minimize the monthly payment. The changing cell is
the monthly payment. The only constraint is that the ending balance in Month 10
equals 0.
Enter the beginning balance in cell B5. You can enter a trial monthly payment in
cell C5 and then copy the monthly payment to the C6:C14 range.
The monthly interest rate (cell C1 is named Rate) is computed in D1 by dividing
the annual rate of 0.08 by 12. The interest paid each month is computed by copying
the rate*B5 formula from cell D5 to D6:D14.
The goal is to minimize the monthly payment (cell C5). Note that the changing cell
is the same as the target cell. The only constraint is that the ending balance for
Month 10 must equal 0. Adding this constraint ensures that the loan is paid off.
After you choose the Simplex LP engine and select the nonnegative variables
option.
Solver calculates a payment of $1,037.03, which matches the amount calculated
by the Excel PMT function.
This model is linear because the target cell equals the changing cell, and the
constraint is created by adding multiples of changing cells.
It should be mentioned that when Solver models involve very large or very small
numbers, Solver sometimes thinks models that are linear are not linear. To avoid
this problem, it is good practice to select Use Automatic Scaling in the Options
dialog box. This should ensure that Solver properly recognizes linear models as
being linear.