Doctor of Engineering (DEng)
| Unit code | DENG801 | ||
| Unit name | Advanced Data Analysis | ||
| Assessment # | 3 | ||
| Paper # | A | ||
| Version # | 1 | ||
| Created by | Vinnu Madhav | Date | 8 Apr 2022 |
| Reviewed by | Yuanyuan Fan | Date | 8 Apr 2022 |
Assessment Instructions:
- Please familiarise yourself with the EIT Academic Honesty and Misconduct Policy, in order to understand your requirements and responsibilities as a student of EIT.
- Please refer to our Assessment, Moderation and Student Progress Procedure for information relating to extensions. Extension requests should be submitted to your LSO at least 3 days prior to the due date.
- Assessments submitted via email will not be accepted.
- Assessments must be submitted through Turn-it-in (unless otherwise stated). Your submission must:
- Be a single document (Word or PDF only)
- Include at least 20 words of machine-readable text, and
- Not exceed 10MB.
- You must use the provided assessment cover page available on your Moodle student homepage. Submissions without a cover page will not be accepted.
- You must correctly title your document/s. For example:
UNIT#_ASSESSMENT#_YOURNAME_DATE
E.g. ME501_Assessment2_SteveMackay_01Aug2019
- You must reference all content used from other sources including course materials, slides, diagrams, etc. Do not directly copy and paste from course materials or any other resources. Refer to the referencing section of the EIT eLibrary on Moodle for referencing guides.
- It is your responsibility to check that you have submitted the correct file, as revised submissions are not permitted after the due date and time.
Important note: Failure to adhere to the above may result in academic penalties. Please refer to the unit outline or EIT Policies and Procedures for further information.
Please complete your answers on the assessment cover page document available on Moodle.
Clearly label your question numbers (there is no need to copy the full question over). Include all working out.
Instructions:
- Students may use a scientific calculator, MS Excel, XLSTSAT, Minitab, Statistica, SPSS, Design-Expert, or any other statistical software for solving and plotting.
Case Study Description:
The injection moulding process is used for manufacturing plastics. It is decided to apply the Design of Experiments (DOE) concept to improve product quality by reducing defect “excessive flash”.
Four important process parameters which believed to be affecting flash formation are: pressure (A) [high level 30 bar, low level 10 bar], time (B) [high level 5 sec, low level 1 sec], speed (C) [high level 50 mm/s, low level 12 mm/s], and RPM of the screw(D) [high 200, low 100].
Part I: 24 Full Factorial Design (30 marks)
- Create the design matrix for 24 factorial design. (2 marks) The following data may be used for creating the 24 factorial design matrix table:
| Run Label | Excess Flash (mm) |
| (1) | 0.22 |
| A | 6.18 |
| b | 0 |
| Ab | 5.91 |
| C | 6.6 |
| ac | 6.05 |
| bc | 6.76 |
| abc | 8.65 |
| D | 0.46 |
| ad | 5.06 |
| bd | 0.55 |
| abd | 4.84 |
| cd | 11.55 |
| acd | 9.9 |
| bcd | 9.9 |
| abcd | 9.9 |
- Calculate the main effects and their percentage contribution. Also find the sum of squares for each case. (4 marks)
- Can you apply F-test for this case study application? If not, suggest a potential solution.
(2 marks)
- Draw the normal probability plot of the effects. (3 marks)
- Make comments about significant factors and interactions based on the observations from normal probability plot. (2 marks)
- Plot the main effect of significant factors and interactions. (5 marks)
- Identify which factors are not important and remove them. After removal of unimportant factors, redesign the experiments with new factors. Prepare a table of factor effect estimate, sum of squares and run ANOVA test. (5 marks)
- Prepare a regression model that represent predicted or best fit value for flash formation. Include only significant factor and interactions. (4 marks)
- Check the model adequacy by normal probability plot of residuals and residuals versus predicted value of flash size (3 marks)
Part II: 24-1 Fractional Factorial Design (15 marks)
- Prepare a fractional factorial design table and calculate the residuals. (4 marks)
- Estimate the effects and prepare the aliased structure. (3 marks)
- Plot the normal probability of main effects. (3 marks)
- Based on the above analysis, build a predictive regression model for the flash size. (5 marks)
Part III: PowerPoint Presentation (10 marks)
Put together a summary of the entire DOE procedure followed for both full factorial and fractional factorial design and comment which one is better. Make a PowerPoint presentation of the entire solution procedure for both full factorial and fractional factorial design. Send your recorded presentation using the following link.
Deliver a 10-minute presentation The requirements are as follows:
- Record your 10-minute presentation and save your recording in the folder using this link: Assessment 3 Presentation Recordings
- Introduce yourself and explain the question and objectives briefly (2 minute)
- Start explaining your solutions (8 minutes)
Presentation Grading Rubric:
| Assessment Criteria | Out of: | Score: |
| Contents (Topic/Question, Student name, Objectives) | 1 | |
| Bullet points – not sentences | 1 | |
| Diagrams and tables (Visible size, Labelling) | 1 | |
| Slide quality (slide numbering, visible font size) | 1 | |
| No spelling/grammar mistakes | 1 |
| Clear information and explanation | 2 | |
| Voice modulation – not monotone | 2 | |
| Time management – completes within given timeframe | 1 | |
| Total: | 10 |
END OF ASSESSMENT
Get expert help for DENG801 Advanced Data Analysis and many more. 24X7 help, plag free solution. Order online now!
