Skip to content

Statistical Engineering - Special course - June 12th to 14th

Date:

June 12th to 14th.

See detailed timetable below.

Classroom:

S01

Instructors

Coordinators:

Xavier Tort-Martorell

És Dr. enginyer industrial per la UPC, màster en Estadística Industrial per la Universitat de Wisconsin (EUA) i professor del Departament d'Estadística i Investigació Operativa de la UPC. Anteriorment ha desenvolupat la seva activitat professional en diverses consultories internacionals, sempre en l'àrea de la qualitat i la millora. Ha assessorat nombroses empreses nacionals i internacionals i ha estat avaluador del Premi Europeu de la Qualitat. És coautor de diversos llibres i de nombrosos articles publicats en revistes especialitzades sobre l'enginyeria de la qualitat.

Lluís Marco

És Dr. enginyer industrial i professor del Departament d'Estadística i Investigació Operativa de la UPC. És expert en tècniques i metodologies per a la millora de la qualitat, ha participat en el disseny i en la impartició dels cursos de formació sobre Six Sigma realitzats per la UPC, ha impartit cursos i ha assessorat a empreses sobre aquests temes a Espanya i Amèrica Llatina . Una altra de les seves àrees d'investigació és l'ús de mètodes estadístics en el disseny emocional de productes.

Presenters:

  • Ronald Does (University of Amsterdam)
  • Roger Hoerl (Union College)
  • Bart de Ketelaere (KU Leuven)
  • Volker Kraft and Ian Cox (JMP)
  • Murat Kulahci (DTU, Technical university of Denmark)
  • Geoff Vining (Virginia Tech)

Discussants:

  • Antje Christensen (Novo Nordisk)
  • Alberto Ferrer (Universitat Politècnica de Valencia)
  • Laura Freeman (Institute for Defense Analysis)
  • Luisa Puerto (Maxam)

Language

English

Description

Engineering is defined (Webster's dictionary) as: "the application of science and mathematics by which the properties of matter and the sources of energy in nature are made useful to people". Engineering is thus the way to make science useful to people. This is the central idea of Statistical Engineering, to make the statistical science useful to people. Therefore, the course is not about how to apply statistics to engineering issues, in the same way as chemical engineering is not about how to apply chemical science to engineering issues. Statistical Engineering is an emerging field aimed at systematize (provide structure and organization) the way to make statistical science useful. The first steps towards this end are directed to study how to solve complex problems using data. As the problems to be solved are complex, the solution often requires the integration of different approaches: statistical concepts and tools, but also other knowledge from relevant disciplines.

The course main objective is discussing and advancing the emerging discipline of statistical engineering.
The following topics will be covered:

  • What is statistical engineering, and why it is the right time for it.
  • Why statistical engineering is different from applied statistics or data science.
  • How the idea originated and evolved.
  • Case studies of statistical engineering.
  • Statistical engineering in Europe: what is in it for European professionals and companies.

More information can be found at:
www.enbis.org
www.isea-change.org

Course goals

At the end of the course, the attendees will understand the principles of Statistical Engineering: combine statistics, data science and IT technologies to solve large, complex and often unstructured problems. They will Know the logic behind this principles and how are implemented through the six phases of Statistical Engineering problem solving. This will be illustrated by real cases in different types of settings.
Additionally the course will provide the attendees with an international vision of pros and cons given by an extraordinary selection of experts in the field coming both, from industry and academia and representing 6 different countries.

Course contents

Wednesday. June 12th (9:30 to 13:00)

  • Course introduction: objectives, course work and evaluation system
  • Background to improvement methodologies: Juran's Breakthrough and Six Sigma DMAIC
  • Introduction to Statistical Engineering
TimeThursday. June 13Friday. June 14
9:00 to 10:45 1.1
What is SE, why it is needed, why it is the right time for it. What’s in it for statisticians. What are the benefits for everybody: society, companies, etc. Implications for statistical training

Primary Speaker:  Ronald Does (45 minutes)
First Discussant:  Laura Freeman (20 minutes)
Second Discussant: Antje Christensen (20 minutes)
Floor Discussion (20 minutes)

2.1

 

How SE is different from: Applied statistics and data science. Simple and well-known examples of SE like DMAIC (and why it is not enough), or clinical trials. How the idea originated and evolved. Lessons from examples and how they are the basis for a new science. 

 

Primary Speaker:  Geoff Vining (45 minutes)

First Discussant:  Alberto Ferrer (20 minutes)

Second Discussant:  Luisa Puerto (20 minutes)

Floor Discussion (20 minutes)

10:45 to 11:15

Coffee Break

 

11:15 to 12:45

1.2

SE cases/ examples

Speaker 1: Murat Kulahci (35 minutes)

Speaker 2: Bart de Ketelaere (35 minutes)

Floor Discussion (20 minutes)

 

 

 

2.2

What has ISEA done so far and what is planning to do. Development of the SE Book of Knowledge. 

Primary Speaker:  Roger Hoerl (40 minutes)

Is SE interesting for European businesses? What can be done to involve European professionals and companies, etc.

Round table (50 minutes)

Participants: Selected plenary speakers and discussants

12:45 to 13:15

What’s the Role of Software in Statistical Engineering?

Ian Cox and Volker Kraft

SAS Institute – JMP Division

Floor discussion (30 minutes)

Facilitator:  Xavier Tort or Lluis Marco

 

 

13:15 to 14:30

Lunch
14:30 to 18:00

Contributed Sessions

2 sessions with 3 speakers each in parallel.  (Total of 12 talks).

Coffee break around 16:00-16:30.

20:30 Networking dinner

Prerequisites

There are no prerequisites.
Interest in applying statistical methods to solve relevant problems and data based decision making
Having practical experience on using data to solve complex problems is a good background.
MESIO's course "Statistics for Business Management" is a good reference and provides a good background.

Targeted at

  • Students and academicians interested in the use of statistics, analytics, data science, IT and the like in solving relevant practical problems. Sometimes this is considered more an art than a science, something that can only be learned by experience. The course is targeted at people interested in participating and contributing in the thrilling development of a new discipline aimed at solving complex, unstructured problems through data in an structured and systematized way.
  • Professionals interested in quality and productivity improvement, data based decision making, sis sigma and improvement methodologies in general.
  • People interested in the application of statistical methodologies to solve practical, relevant problems

Evaluation

In addition to attending the presentations and participating at the discussions of the ENBIS Spring Meeting students will be asked to:

  1. Watch the video: "Statistical Engineering: A Holistic Problem-Solving Approach". With Roger Hoerl
  2. Read the article: "How is Statistical Engineering Different from Data Science?" by Allison Jones-Farmer and Roger
  3. Prepare a report answering the following questions:
  • What are the aims of Statistical Engineering?
  • Is it reasonable to center SE in solving complex problems or should be extended to other issues? If yes which ones?
  • Do you think that the 6 phases are reasonable? Are they sufficiently devoted?
  • What must be done to extend SE to non-statisticians?
  • What is your prognosis for SE future?

Computer class or student's laptop?

Students do not need a computer to follow the course