Six Sigma and LPE

Discuss, Discover, Learn, and Share. Feel free to share information.

Moderators: orrb, saw22

Post Reply
Kasey Wertheim
Posts: 161
Joined: Sun Jul 10, 2005 6:55 am

Six Sigma and LPE

Post by Kasey Wertheim »

Has anyone coordinated, experienced, or checked in to bringing latent print examination processes under Six-Sigma compliance? Usually, this is a Quality Assurance manager's job in coordination with examiners in the section - there is an event where business processes are mapped out, waste is defined, and procedures are trimmed to achieve better effectiveness (accuracy and efficiency). Usually, events are coordinated by a Six-Sigma black belt or at least a yellow belt. If you are willing to discuss this with me for a few minutes, please provide details or your phone number via e-mail: kasey.wertheim@us.army.mil. If anyone has thoughts to share with all, feel free to reply via forum post. Thanks!
Michele
Posts: 384
Joined: Tue Dec 06, 2005 10:40 am

Re: Six Sigma and LPE

Post by Michele »

Kasey,

You’re contemplating efficiency in government work… are you high? Sorry, I mean…it’s great to see someone thinking outside the box.

Here’s my uneducated opinion (no belt):

I think most agencies could streamline processes and /or use their money better but six-sigma services don’t come cheap (or any other type of assessment services)!! Even more important than having someone recommend improvements, management has to be willing to make those changes and maintain the changes. I think that agencies need to go into something like this with the idea that it’s a continuous process of refining procedures, not simply a quick fix that can be made if you throw enough money at it. Our agency has gone through process mapping a few times since I’ve been here (none where strictly six-sigma).

I think one way that agencies can make improvements without a huge investment is to talk to those people that do the work and take their suggestions seriously.

As an example, let’s look at LAPD. Here’s an agency that has over 70 examiners (that’s what I’ve heard) and an audit division (far more resources than most agencies). With these resources, they’ve decided that the only way to improve performance and accuracy is to spend about $400,000 to hire someone else to assess the unit and make suggestions to the process and workflow. I would bet that they have several competent people out of those 70 people that could make the same suggestions that an outside company would make but management has the perception that the cost must be related to the quality. Sometimes the best suggestions are free, we just need to be willing to listen to them.
Michele
The best way to escape from a problem is to solve it. Alan Saporta
There is nothing so useless as doing efficiently that which should not be done at all. Peter Drucker
(Applies to a full A prior to C and blind verification)
Gerald Clough
Posts: 557
Joined: Wed Jul 06, 2005 6:27 am
Location: Lockhart, Texas
Contact:

Re: Six Sigma and LPE

Post by Gerald Clough »

Heretic! You'll surely burn.

In the organizational environment, what's free is deemed questionable, while what you pay $400,000 for just must be worth doing. Donchaknow.
"Nothing has any value, unless you know you can give it up."
Ann Horsman
Posts: 152
Joined: Thu May 17, 2007 8:01 am
Location: VERMONT
Contact:

Re: Six Sigma and LPE

Post by Ann Horsman »

+1

Michele, right on.
~Ann

Image
Veritas vos liberabit

"...but no prints can come from fingers
if machines become our hands"


Jack Johnson - The Horizon Has Been Defeated
Ernie Hamm
Posts: 214
Joined: Sun Jan 22, 2006 10:24 am
Location: Fleming Island, Florida
Contact:

Re: Six Sigma and LPE

Post by Ernie Hamm »

A check with an ASTM associate resulted in the following: It is an ISO term stating that 95% of the product will be within +/- 3 sigma of the required measurement value. and to have six-sigma compliance you have to have something to measure; a dimension, a mass, a time. a performance. etc. You need a set of values on which you can do statistical analysis.. It seems the "belts" refer to the degree of training/knowledge in the area, much like the martial arts (there is even a Six Sigma green belt along with the Black and Yellow).

The "six' references the +/- 3 (3 below the standard and 3 above the standard = six). Using the ISO term of 95% could mean your latent print performance/accurancy can be from 92% to 98% and you would be "six sigma compliance". That is not very good in the practice of latent print identification.

Back to the tar pit.
Michele
Posts: 384
Joined: Tue Dec 06, 2005 10:40 am

Re: Six Sigma and LPE

Post by Michele »

My understanding (which shouldn’t be given a whole lot of weight) is that the six sigma standard is actually an extremely high standard. The +/- 3 is referring to standard deviations, not percentage points. Statistically speaking, this is about 3.4 defects per million opportunities for a defect. My previous post was really talking about generalities in refining procedures but six sigma is a specific way measure if a certain procedure has extremely high (or extremely accurate) results.

There are several ways to refine procedures to make a company or agency more efficient. You can diminish waste in a system (like cut unnecessary steps being done), you can increase production (by improving morale or by making employees produce more), or you can reduce unwanted results. In manufacturing, ‘unwanted results’ may be called ‘errors in production’ or ‘defects’. It’s important to reduce these because defects are considered a waste and therefore inefficient. Reducing defects is the same as reducing variation in a system.

The six sigma standard is really an attempt to reduce variation in any certain procedure to a very low level. This is done by clearly stating what is expected from a procedure and then measuring to see how close it comes to what is desired. This procedure would comply with the six sigma standard if the defects are within the 3.4 defects per million range. ‘Defects’ are considered to be anything that is outside of the standard that is set (the standard being the desired results).

For instance, a company (or agency) may want to measure the false negative results in an AFIS computer system. Before you measure the false negative rate you’d have to state what is considered to be a valid false negative. This could mean that you’d have to say how many candidates will be looked at (which might depend on how many minutia are in the print). And there might be other things that need to be looked at in determining what constitutes a false negative. Once the standard is set then you have to test the system to find the false negative rate. It may qualify as meeting a six sigma standard if the false negative rate is under the 3.4 defects per million searches. There are other standards too…five sigma, four sigma, three sigma… all of these have more defects per million than the six sigma standard (trials that don’t fall within the desired standard). Six sigma is considered by most to be the highest standard. In order to know how well any procedure performs you have to be able to accurately describe the expectations, accurately measure the results, and accurately record or report them. These are the main points to improve a process that you want to meet the six sigma standard.

Someone could also measure the false positive rate in a lights out system, or maybe measure a variety of non AFIS related procedures. Usually the key is to measure procedures that are done in extremely high volumes (that's why this is common in manufacturing...because they produce high volumes of items). Procedures that aren’t done in high numbers usually can’t be measured within the six sigma standards.

As far as belts, this usually refers to a person with experience. It may also refer to projects but I’m not aware of that. Yellow doesn’t seem very common. Some companies don’t even recognize yellow belts. Green belt, Black belt, and Master Black belt seem more standard. People (expert six sigma consultants) gain the better belts by meeting objectives that are set. This includes the number of projects they’ve worked on, the types of projects, the number of projects that have progressed to six sigma, etc.

At least that’s my understanding of it.

Kasey – are you looking for something specific about six sigma?
Michele
The best way to escape from a problem is to solve it. Alan Saporta
There is nothing so useless as doing efficiently that which should not be done at all. Peter Drucker
(Applies to a full A prior to C and blind verification)
Kasey Wertheim
Posts: 161
Joined: Sun Jul 10, 2005 6:55 am

Re: Six Sigma and LPE

Post by Kasey Wertheim »

Nothing specific - just looking to see whether anyone has actually implemented this level of QA in a latent print operation. I appreciate all the detail, and if anyone logs on this thread in the future who has experienced this, please feel free to chime in. Thanks all,
-Kasey
Royce D. Wilson
Posts: 7
Joined: Tue Dec 12, 2006 6:56 am
Location: Tampa, Fl

Re: Six Sigma and LPE

Post by Royce D. Wilson »

This is an interesting concept that I think will work. While working on my Masters degree in Management, Leadership, and Organizational Effectiveness, I was fortunate enough to have an opportunity to study Six Sigma, and think it can be reduced to a definition of a philosophy of measurement and continuous improvement. While this is an over-simplification of a complex business approach, I believe it will work for this purpose. I think the key to success in applying a six sigma approach to the latent print world is the development of an in-depth flow chart of the processes used in your agency. Every aspect of every area of the various processes in the flow chart should be examined for flaws, weaknesses, waste, or simply opportunities for improvement. A strategy or strategies designed to achieve improvement must be developed and implemented and the outcomes measured. these measurements should lead to the development and implementation of further improvement strategies. Central to this approach is an assumption that a clearly defined mission exists and that all processes are in place explicitly to achieve this mission.

I don't know that a six sigma level of performance (a success rate of 99.99966% --- or, 3.4 defects/failures per million) can be achieved in all areas of the latent print processes; however, I do know that it can and should be achieved in the identification / individualization aspect. But when you understand that included in the processes you have filing, the retrieval of known fingerprints, missed identifications, returning phone messages and e-mails, and on and on, achieving a six sigma level of performance is difficult. However, you should also remember that most companies today perform far below six sigma, and that any move or shift from where they are now toward six sigma is a tremendous improvement.

If you would like to discuss this further, I can be reached at (813) 247-0500.
Royce D. "Rayce" Wilson
Hillsborough County Sheriff's Office
Tampa, Fl
Post Reply