Critical success factors for efficient use of statistical techniques

I work in a small medical device company and have been asked to give a presentation on how statistical techniques can be best used by the company.

I’m going to approach it in terms of what are the critical success factors for successful use.

So far I have the following:

Competence of staff
Correct choice of technique for each application
Good culture of statistical thinking
Management seeing strategic value of stats tools
Documented procedures to aid in use of tools
Software
Availability of data
Analysis of distribution of data, and after reliability of results (OC curve etc)

Am I thinking along the right lines or missing loads.

Any thoughts would be greatly appreciated!

Common Sense !!!!!!!!

irelandoem said:

I work in a small medical device company and have been asked to give a presentation on how statistical techniques can be best used by the company.

I’m going to approach it in terms of what are the critical success factors for successful use.

So far I have the following:

Competence of staff
Correct choice of technique for each application
Good culture of statistical thinking
Management seeing strategic value of stats tools
Documented procedures to aid in use of tools
Software
Availability of data
Analysis of distribution of data, and after reliability of results (OC curve etc)

Am I thinking along the right lines or missing loads.

Any thoughts would be greatly appreciated!

Click to expand...

Performance measures could be another good one.

It is far easier to disseminate and sustain the use of statistics if you can demonstrate a measureable benefit in terms of cost or quality improvement.

Here is a checklist of the steps you should consider:

1) Determine and make a list all of your process variables

2) Verify which of the variables make the biggest impact on your process variability (design of experiments is a handy tool for this)

3) For the most significant variables, perform process capability

4) From the understanding of the process capability study (distribution, adjustments, etc.) determine the proper control methodology. (This is the point you may want to determine if you have the correct software)

5) Implement the control methodology with properly trained personnel

6) Monitor the results - after more data is collected, ensure your original theories still hold true or readjust (PDCA)

thanks guys!

theres a lot of info out there on the methodology for applying individual tools.

thats a good structured approach to take, thanks bob and prototyper I have added KPI's to the list.

I'm structuring the approach in a fishbone diagram so I have

man
qualifications
training
competence in use of tools
belief in value of tools

method
which tools used
where tools used
used pro actively
used reactively

environment
integration with quality mgt system
culture of use of stats
strategic value of stats
documented procedure

machine
software packages

material
data readily available
cost of generating data

measurement
underlying distribution
quantify reliability of results
key performance indicators

Any ideas on considerations that should be added or changed?

Interesting concept, assembling these issues into a fishbone. What is the defined output of these effects? "Efficient use of statistical techniques"? Remember, that is a very broad term. Do your really mean it, or are you focusing on Statistical Process Control?

bobdoering said:

Interesting concept, assembling these issues into a fishbone. What is the defined output of these effects? "Efficient use of statistical techniques"? Remember, that is a very broad term. Do your really mean it, or are you focusing on Statistical Process Control?

Click to expand...

Output is exactly that "Efficient use of statistical techniques"

I understand it is a very broad term, thats why I thought structuring in fishbone would help identify areas. I’m looking at all statistical techniques right from pareto and box plots to taguchi and response surface methodology.

[FONT=&quot]It is also part of work I am doing for a university project, hence looking at such a broad (academic) field.[/FONT]

Where does risk control factor into your evaluation?

For us risk drives quite a bit of the discussion.

You want to concentrate your efforts on the items that pose a risk to your customer and your business.

hi guys, i've put together the attached ishikawa diagram.
Anybody got any thoughts on it?

irelandoem said:

hi guys, i've put together the attached ishikawa diagram.
Anybody got any thoughts on it?

Click to expand...

Can you clarify what you're trying to accomplish with it?

Stijloor.