Part-1:
A leading precast manufacturer has decided to replace a chemical admixture which has been used for more than a decade with a new chemical admixture to control early set and accelerate strength development. The advantage claimed for the new admixture is that it has excellent compatibility with air entraining admixture used in some precast products. However, the plant manager has suggested completing an experimental trial so that the benefits of the new admixture can be established before making changing to the materials used.
Task 1
Prepare an outline of the experimental design you will suggest to the plant manager. This outline should contain research rationale, aims and objectives, research methodology, nature of data analysis and interpretation, and possible outcomes/conclusions to be drawn. [The research methodology should contain details of materials to be used, tests to be carried out and experimental controls to be established so that meaningful data can be obtained. The nature of data analysis and interpretation should deliver conclusive findings to address the aims and objectives.]
Task 2
Identify learning resources that you will consult for understanding the topic of research, materials, test methods and data analysis. Prepare a list of references which can be consulted for completing this task.
Part -2:
Task #01:
- Please see "Design of Experiment Methods" presentation - 2nd slide on page 1.
- What should you be measuring? Define the dependent variable(s).
- How are you going to measure air entrainment? How are you going to measure the influence of these two admixtures on other properties of interest, e.g. setting time, early compressive strength. Research available/variety of tests to be used and decide on most appropriate one(s).
Justify your choice.
- You have to start designing the experiment you want to do. You have to decide on type of experiment. You may want to propose a simple experiment comparing mean values of the two factors where there is only one level or may decide to go for a more complicated experiment where you want to investigate different levels of the two factors.
- Plan your experiment. Decide on how many repetitions you will need. Decide on cost implications.
- Depending on the test there may be requirements on the standard deviation and further tests may be required to prove this assumption.
- List the assumptions for which the analysis is valid for, e.g., for normality, standard deviation, etc. Please see replies to recent queries.
- The test may not necessarily be a test of variance. Select most appropriate type of test.
- Note that all but parametric testing requires data to be normally distributed.
- Explain the analysis of the particular experiment chosen. At what level is one better that the other. Significance depends on the application. Discuss the level of significance.
Task #02:
Task #02 only requires listing of sources of information you used to make the above suggestions.
Task #01:
Please see "Design of Experiment Methods" presentation - Pages 14 to 15. The slides show you how to change 2k factorial design to central composite design.
Task #02:
Please see Group Assignment #02 presentation.
Task #03:
Central Composite Design - What will the equations for the regression analysis look like? Please see "Design of Experiment Methods" presentation - Pages 14 to 15.
Task #04:
Please see "Design of Experiment Methods" presentation - Pages 13 to 15
Task #05:
Explain what information you can get out of Table 5.
Task #06:
Explain what information you can get out of Table 6.
Task #07:
Either plot the equations on page 3 or otherwise try to highlight the trends in Table 4 on page 4.
Attachment:- Assignment.rar