Lean Manufacturing Six Sigma Forum

Bienvenido. Welcome. Bienvenu.


Temas únicos de Lean Six Sigma. Foro creado en el 2010

Últimos temas

DataFeast By Quantum TC

Numbers role in Lean Manufacturing.


Mensajes : 105
Puntos : 308
Fecha de inscripción : 30/12/2010

Numbers role in Lean Manufacturing.

Mensaje  Admin el Dom Jul 03, 2016 9:36 am

There is no better way to start this English section of this six years-old spanish speaking Lean Six Sigma forum than with this Taiichi Ohno citation.

I came across this citation a few days ago and it reflects perfectly the necessary union of two complex but completely complementary and integrable world: lean manufacturing and six sigma. Six sigma is mostly quantitative. But numbers are the effect of events. One needs to know and understand the event mode to improve numbers.

Say you have a process that delivers a product at the rhythm of one every 15 seconds (process cycle time is 15 seconds).  And assume this is high consumer product and works two shifts of 8 hours each.  At the end of 10 working days you have accumulated 38,400 products. Each product say have 10 defects opportunities. That is you have on hand 384,000 opportunities to fail. No matter if you have failed once or more or if you haven't failed at all, this data is rich for the future. But it is purely data.

The transformation of data from any transactional system into unstructured informations (bringing relevant data fields together from a query system) then into structured information (descriptive statistics like mean, range, quartiles, standard deviation, skewness, and even simple comparison)  for its direct exploration, analysis, o alerts, and then it's conversion into knowledge (understanding the facts behinds the numbers by direct observation, seeing and listening of the process), is the fundamentals for any organization to start developing its Business Intelligence.

If we don't understand the numbers generated by our processes, how good are we for the process? Could we manage the process? Could we improve the process? The answers to all are NO, if we are not good at the numbers of the process. We don't have to be an engineer, but we do have to be able to understand the process numbers no matter the process, no matter our background.  

So the first step into having parties involved in a process understand the numbers behind the process, is to make the numbers visible, (visual management in lean manufacturing), in the form of structured informations: reports, metrics,  and even pictures.  Yes if you have learnt to see, pictures have numbers all over them in the form of informations. By doing this you will be generating enough interest and the most curious will try to deepen their understanding especially if there is a lean culture of Gemba walks and kata management questioning the numbers.  Not having a visual management system in the Gemba is to eliminate any opportunity of breaking the comfort zone underlying the numbers.

However what is the purpose, if any, to manage pure numbers and post informations on the wall, if one does not go and see (genchy gembutsu) for one self, the facts behind the event in order to fully grasp all the aspects of the process event?  Not doing it would be a waste of time recollecting data. Not doing it would a waste of money in some complex or simple measurement system. It would be just data manipulation and nothing more. Looking at pure numbers or even looking at information and not doing anything about is is the worse of all as our master Taiichi told us.

Data must give informations. Informations must give knowledge. If we don't understand the numbers we will not have the right informations. If we don't post the right informations and generate the curiosity of the how we will not have knowledge of the business. This is all about Business Intelligence. Indeed if you really look deeply into the Toyota Production System, TPS, you will see that it is a business intelligence driving system. No wonder why TPS is also referred to as Thinking People System.

Administrador del Foro
Quantum TC
Dominican Republic

    Fecha y hora actual: Mar Nov 13, 2018 11:39 am