Industrial Statistics III

Go to this course in the University Course Planner.

Description

The course is concerned with the control and improvement of processes. Most of the examples are taken from engineering processes, but the methods are widely applicable in other areas such as commerce and medicine. The course is in three parts, related by the theme of process improvement. The first part is statistical process control, which can be defined by three questions: is the process stable; is the process capable of meeting the specification; and how can the process be monitored to check it remains in its stable state? The second part is an introduction to the design and analysis of experiments, with the intention of improving the process so that it remains competitive in world markets. The third part covers the life distributions of individual components, the reliability of systems that cannot be repaired and the analysis of repairable systems.

Objective

To introduce students to statistical methods for monitoring and improving processes. Students should understand statistical process control and the design of experiments, and be able to implement these in the open source software R. Students should be able to carry out reliability analysis using R. Students should be able to set up models for the reliability of non-repairable and repairable systems.

Content

Components of variance; autocorrelation function. Process control indices. Statistical process control charts: Shewhart mean and range, C-chart, P-chart, cusum chart, moving average charts, exponentially weighted moving average chart. Multiple regression. Indicator variables for qualitative factors. Factorial and fractional factorial designs, central composite designs and evolutionary operation. Taguchi style experiments. Weibull models for component reliability, censored data. Kaplan-Meier estimator Structure functions for system reliability. Markov models for repairable systems.

Andrew Metcalfe Lecturer for this course

Delivery

Two 1 hour lectures per week and a 1 hour practical each fortnight.

Assessment

2 hour examination (85%) and 4 class exercises (15%)

Graduate attributes

• Applying mathematical knowledge (1)
• Interpreting data and drawing conclusions (2)
• Formulating mathematical problems (3)
• Problem solving skills (4)
• Flexibility in working environment (5)

Linkage past

MATHS 1012 (Pass Div I) or MATHS 2004 (Pass Div I), one of STATS 1000 (Pass Div I), STATS 1004 (Pass Div 1), STATS 2004 (Pass), APP MTH 2009 (Pass), STATS 2001 (Pass)

Linkage present

No present linkages have been noted.

Linkage future

This course is not recorded as prequisite for other courses.

Restrictions

None.

Recommended text

Modern Industrial Statistics –R.S. Kenett and S. Zacks [1998,Duxbury]. References: Statistical Methods for Engineers (2E)– G.G. Vining and S. Kowalski [2006,Thomson] Reliability and Risk Assessment –J.D.Andrews and T.R. Moss [1993,Longman]. Practical Methods for Reliability Data Analysis – J.I. Ansell and M.J. Phillips[1994, Oxford University Press]. Out of the Crisis – W.E.Deming [1986, MIT]. Probability and Statistics for Engineering and the Sciences (5E)– J.L. Devore [2000,Duxbury]. Design and Analysis of Experiments (6E) – D.C. Montgomery [2005, Wiley] Statistical Quality Control (2E) – D.C. Montgomery [1991, Wiley] Probability and Statistics for Engineers and Scientists(2E) – R.E. Walpole and R.H. Myers [1978,Macmillan].