Final non presentation individual project business modeling and applications

Untitled UNIVERSITY OF ECONOMICS HO CHI MINH CITY FACULTY OF INTERNATIONAL BUSINESS MARKETING FINAL NON PRESENTATION INDIVIDUAL PROJECT BUSINESS MODELING AND APPLICATIONS  oOo  Name of Lecturer Nguy[.]

UNIVERSITY OF ECONOMICS HO CHI MINH CITY

FACULTY OF INTERNATIONAL BUSINESS - MARKETING

FINAL NON-PRESENTATION INDIVIDUAL PROJECT

BUSINESS MODELING AND APPLICATIONS

 oOo 

Name of Lecturer: Nguyen Thi Hong Thu

Subject code: 22C1BUS50321204

Student ID: 31211020894

Student name: Truong Dang Hong Duong

Date: 22 December, 2022

University of Economics Ho Chi Minh city Faculty of International Business - Marketing

Business Modeling and Applications

Final Non-Presentation Individual Project

2

Endorsement

The author of this essay, Truong Dang Hong Duong, asserts that it was a personal undertaking overseen

by my course instructor, Ms Nguyen Thi Hong Nhung

All of the research and writing in this document is based on my own knowledge that I learned from the instructor I certify that neither other people's nor other organizations' work was used to create this publication The information in this post was individually gathered, examined, and summarized using Excel Solver and QM for Windows

If any fraud is discovered in this article notwithstanding the claims made above, I shall be entirely liable for its content

Thursday, 22 December 2022

Author (Signed)

Truong Dang Hong Duong

Lecturer’s Comment

MAIN CONTENT

1 Linear Programming (05 Pts)

Imagine a similar problem and analyze it:

HD is a company that creates public services for its community It has six consultants and eight clients on board for the project Because the consultants have varying technical abilities and experience, the company charges different hourly rates for its services Furthermore, some consultants' skills are better suited for certain projects than others, and clients may prefer one consultant over another A consultant's suitability for a project is graded on a 5-point scale, with 1 being the worst and 5 being the best The table below shows each consultant's rating for each project, as well as the hours available for each consultant, as well as the contracted hours and maximum budget for each project:

Project Consultant Hourly

to maximize the hours that are specified and rated The points the higher the better (from 1 to 5)

The Available hours as a table below:

The same with Project hours:

And The Contract budget

8

The number of hours that they consult per project: xiy ≥ 0

If I apply the mathematical models to solve this problem:

First, I need to define the variables as shown in the table below:

A1, A2, A8 Hours that Consultant A is assigned to do Project 1, 2, , 8

B1, B2, B8 Hours that Consultant A is assigned to do Project 1, 2, , 8

C1, C2, C8 Hours that Consultant A is assigned to do Project 1, 2, , 8

D1, D2, D8 Hours that Consultant A is assigned to do Project 1, 2, , 8

E1, E2, E8 Hours that Consultant A is assigned to do Project 1, 2, , 8

Downloaded by hây hay (vuchinhhp3@gmail.com)

F1, F2, F8 Hours that Consultant A is assigned to do Project 1, 2, , 8

The following formula can be used to calculate the objective or total suitability:

3A1+3A2+5A3+5A4+3A5+3A6+3A7+3A8+3B1+3B2+2B3+5B4+5B5+5B6+3B7+3B8+2C1+C2+3C3+3C4+2C5+C6+5C7+3C8+D1+3D2+D3+D4+2D5+2D6+5D7+D8+3E1+E2+E3+2E4+2E5+3E6+3E7+3E8+4F1+5F2+3F3+2F4+3F5+4F6+3F7+3F8

 Solve this problem using QM

I select Module, then Linear Programming Options After choosing the correct type of the module, I define the number of constraints and the number of variables

I have three types of constraints: The available hours of each consultant, The hours of each project and revenue from each project:

- Each consultant’s availability: I have six consultants named A, B, C, D, E and F So that I also have 6 available hours of each person (Available hours of A, Available hours of B,… E)

- The hours of each project: I have 8 projects, which leads to 8 constraints: Hours of Project 1,

Add 6 consultants to 8 constraints of hours and 8 constraints of revenue, I have 22 Constraints total Column 1 will then show that consultants A work on project 1

Columns 2 will express that consultants A do the project 2…

So that I have 6 consultants with 8 projects The total Columns are called Variables can calculate by: 6*8 = 48 Variables

Choose the objectives Maximize After that, in the row of Maximize I fill with the numbers (the points)

in the table of the topic

When I input the correct data, I will get the result below:

 Solve this problem using Solver

12

c Create a sensitivity report

By using Solver, I can easily expertize a sensitive report If you want to see more detailed and

completed, access the Excel file attachment

Sensitivity analysis gives you insight into how the optimal solution changes when you change the coefficients of the model After the solver finds a solution, you can generate a sensitivity report

Downloaded by hây hay (vuchinhhp3@gmail.com)

14

d Explain the shadow price, reduced cost and the allowable range (increase and decrease)

Shadow price illustrates:

- The value of each additional unit

- The marginal or additional value of an additional unit of a resource

- Over and above what you pay for resources

Reduced cost depicts:

- How much improvement in each objective function coefficient is required before the related decision variable in the ideal solution assumes a positive value

- It illustrates our potential gains and losses from manufacturing a good we are not currently generating

- The amount by which each additional unit generated would raise the Z value

- Equals: unit contribution (= Cj - opportunity cost)

Allowable Range Decrease is the amount that can be decreased with shadow price staying constant, which demonstrates:

- How much it can decrease without changing variables

Allowable Range Increase is the amount that can be increased with shadow price staying constant, which demonstrates:

- How much it can increase without changing variables

2 Decision Making (03 Pts)

Petrolimex Gas station are soon going to open a new dealership They have 3 offers: from a local gas company, from a provider and from a big gas corporation The success of each type of dealership will depend on how much gasoline is going to be available during the next few years

Fill the profit of each type of dealership, giving the availability of gas data to the following payoff table (unit: million VND) Draw a decision tree to help Petrolimex choose what’s best for the profit

Downloaded by hây hay (vuchinhhp3@gmail.com)

Dealership Gasoline Shortage Gasoline Surplus

To solve this Decision Making problem, I used Decision Tree:

Tree Plan creates formulas for summing cash flows to obtain outcome values and for calculating rollback values to determine the optimal strategy Specifically, it can be explained:

Potrolimex can choose and decide the best way for maximize the revenue and expected salvage value

We use the criterion of Expected Monetary Value (EMV) or also known as Bayesian decision rule From Tree plan decision we also know that the expectation – salvage – value of local gas company is 27000$, provider is 43500$ and corporation is 57000

16

Based on this information, Petrolimex can make decision to offer the local company to open a

new dealership It can be explained that in 3 options, Petrolimex has, the local company reach the highest expectation – salvage – value In other hand, EMV cannot completely guarantee that the decision can be made

a Using averaging forecasting method, calculate the forecast

This method uses all the data points in the time series and simply averages them Thus, the forecast of what the next data point will turn out to be is:

Forecast = Average of all data to date

I have applied this method to solve the problem, which is represented in my Excel file:

Downloaded by hây hay (vuchinhhp3@gmail.com)

b Using 3-month moving average forecasting method (n=3), calculate the forecast

The moving average method forecasts for the next period by taking the average of the n nearest values

in a time series The 3-month moving average method is to take the average result of the data from the last 3 months to find a forecast for the next month

Applying the 3-period moving average method to the data of October, November, and December, we have a forecast of the supermarket's bottled water demand in the next June as follows:

Conclusion: From the data calculated using the 3-period moving average method and the demand in the last 3 months (30000, 20000, and 10000), we have a forecast report for the industry manager In the next June, supermarkets purchased 20,000 bottles of bottled water

I also use Excel to represent this method:

18

c Using last-value forecasting method, calculate the forecast

The last-value forecasting method ignores all the data points in a time series except the last one It then uses this last value as the forecast of what the next data point will turn out to be, so the formula is simply:

Forecast = Last value

Applying this formula and calculating the forecasting error, MAD, MSE we have the result as following:

Downloaded by hây hay (vuchinhhp3@gmail.com)

d Explain 3 methods of forecast Which one is better and more accurate according to you? You can explain however you want

In my perspective, MAD and MSE are common measures of forecast accuracy To find the more accurate forecasting model, forecast with each tool for several periods where the demand outcome is known, and calculate MSE and/or MAD for each The smaller error indicates the better forecast (Time-series forecasting, moderate) So according to the results we got, it is obvious that the 3-period moving average is the best of the three

Furthermore, based on the theory, other two methods seem to embrace some limits:

The averaging forecasting method: The estimate is excellent if the process is entirely stable, i.e., if the assumptions about the underlying model are correct However, frequently, there is skepticism about the persistence of the underlying model over an extended period of time Conditions inevitably change eventually Because of a natural reluctance to use very old data, this procedure generally is limited to young processes

20

The last-value forecasting method: sometimes called the "naive method" because statisticians consider

it naive to use just a sample size of one when additional relevant data are available However, when conditions are changing rapidly, it may be that the last value is the only relevant data point for forecasting the next value under the current conditions Therefore, managers who are anything but naive do occasionally use this method under such circumstances

Downloaded by hây hay (vuchinhhp3@gmail.com)

22

Appendix

Appendix A: List of tables

1 Table of available hours

2 Table of project hours

Downloaded by hây hay (vuchinhhp3@gmail.com)

3 Table of Contract budget

24

4 Table of variables definition

A1, A2, A8 Hours that Consultant A is assigned to do Project 1, 2, , 8

B1, B2, B8 Hours that Consultant A is assigned to do Project 1, 2, , 8

C1, C2, C8 Hours that Consultant A is assigned to do Project 1, 2, , 8

D1, D2, D8 Hours that Consultant A is assigned to do Project 1, 2, , 8

E1, E2, E8 Hours that Consultant A is assigned to do Project 1, 2, , 8

F1, F2, F8 Hours that Consultant A is assigned to do Project 1, 2, , 8

Downloaded by hây hay (vuchinhhp3@gmail.com)

3 Image of sensitivity report

28

4 Image of Decision Tree:

5 Image of Averaging Forecasting method

Downloaded by hây hay (vuchinhhp3@gmail.com)

6 Image of 3-month Moving Forecasting method

7 Image of Last-value Forecasting method

30

_The End _

Downloaded by hây hay (vuchinhhp3@gmail.com)