CETYS UNIVERSITY

Engineering School

Industrial Engineering

Statistical Inference Final Project

Sealability Test

Presented by:

Aida Mariana Martinez 23530
Jose Eduardo Pedrero 25209
Alberto Loyola 26751

Mexicali, BC. November 26,2014

Table of Contents
Objective 3
General information about the company 4
Justification of the problem 5
Parameters 6
Models Used 8
Implementing the models 8
Procedure of the Test Process 8
Registered data 10
Capability Histogram 11
Normal Probability Plot 12
Solutions 13
SPC 13
Justification of Model Used 13
Calculations 13 t Distribution 14
Justification of Model Used 15
Calculations 16
Minitab Data 18
Conclusion 20
References 21

Objective

The objective of this project is to analyze the data from the testing of the sealability from the quality department of MEGAPLAST S.C. Co, with the purpose of calculating and documenting the behavior based on the appliance of the techniques learned in class during the course of Statistical Inference.
General information about the company

MEGAPLAST is a company originated in Mexicali, B.C. dedicated to the fabrication of PVC tarpaulin. Initiated in 1972 in Mexicali, B.C.
The company is now located in 12 states and gives service to all Mexico. It also exports everything to the continent of America, from Canada all the way to Venezuela.
The whole tarpaulin is fabricated in Mexicali, B.C., the only production plant available of the company.
The tarpaulins that are designed and fabricated in MEGAPLAST are made thinking in situations in which tension, resistance, and durability are very important factors. Being used mainly for the covers of trucks, circuses, etc.

Justification of the problem

The data of the sealability samples will be analyzed, sample being made by the quality department for a product.
The quality department of MEGAPLAST doesn’t know the capacity of the control limits of their process, this is the reason why they want to know if the products they are manufacturing follow a normal distribution and in what range does the mean of the process have to be in order to have a 95% confidence interval of the product’s process.
This project is focused on the most sold product, Fortaflex 1043 Tarpaulin. The manager of production, Guadalupe Salazar, provided the database containing the index of resistance manufactured in each tarpaulin.
With the information provided, the behavior of the process and the range of the mean of the sealability of the tarpaulins will be determined.
Parameters

The product is produced beginning with the fabric. The raw material are the reels of thread imported from Korea and knitted to stay as tissues.

The product is finished when the tissue is placed in the plasticised machine and added to fusion rollers.

The sealability is the index of the rupture of the tarpaulin, the finished product, after two fibers are attached through heat.
Actually, there are samples made to determine if the process is controlled, although the limits of the specifications of their clients aren’t the most demanding and it counts only as an inferior limit of their specifications.
With a sample size of 25, this project pretends to determine how the process is being held and calculate in what range of values should the mean of the process be to give a 95% CI; answers that can be obtained with the learnings of the course of Statistical Inference.
Models Used

Statistical Process Control Chart (SPC)
Confidence Interval on the Mean of a Normal Distribution
Implementing the models

Sample size: in this case feature, 25 samples of Fortaflex 1043 Tarpaulin.
Frequency: Every thousand meters of tarpaulin an adhesion test is made.
Who does it: The setup is made by the person in charge of the quality tests.
Lower limit required by the client: 30 kg
Procedure of the Test Process

1- Tarpaulin is marked and cut as shown

2- The tarpaulin is sealed with the hot air equipment.
3- The strips are placed to run the test voltage as shown in the