CS代考 FALL 2022 SANJAY DOMINIK JENA, ESG UQAM

Decision Making Technolgies
INTRODUCTION – FALL 2022 SANJAY DOMINIK JENA, ESG UQAM
© SANJAY DOMINIK JENA, ESG UQAM – PLAN DE COURS MBA 8619

Copyright By PowCoder代写 加微信 powcoder

Three definitions of « analytics » according to INFORMS(*):
A synonym for « statistics » ou « metrics »
A synonym for« data science »
 Quantitative approaches to tackle decision making in an organizational context
(*) INFORMS: Defining analytics: a conceptual framework https://www.informs.org/ORMS-Today/Public-Articles/June-Volume-43-Number-3/Defining-analytics-a-conceptual-framework
© SANJAY DOMINIK JENA, ESG UQAM – MBA8419

Word cloud « Analytics » (Wikipedia) https://www.wordclouds.com/
© SANJAY DOMINIK JENA, ESG UQAM – MBA8419

The shortest path problem: find the fastest path, based on quantifiable / measurable information
© SANJAY DOMINIK JENA, ESG UQAM – MBA8419

Examples: descriptive, predictive and prescriptive business analytics
Descriptive Analytics
What happened?
Predictive Analytics
What will likely happen?
Prescriptive Analytics
How should we act?
How many sales did marketing campaign A generate?
How many sales will marketing campaing B generate?
How should we organize the next marketing campaign to maximize sales?
How much annual return did mutual fund A generate the last 5 years?
How much annual return will mutual fund B generate next year ?
Which mutual funds should be positioned in a portfolio to maximize the expected return?
What were the customer demands for product X in region A throughout the last 12 months?
What customer demand can we expect for products X and Y in region B next month?
How many units of products X and Y should we produce to maximize total sales profit?
Marketing Finance
Production planning
Source: http://gestisoft.com/differences-entre-lintelligence-daffaires-et-lanalyse-predictive/
© SANJAY DOMINIK JENA, ESG UQAM – MBA8419

Operations Research (OR)
Descriptive Predictive Prescriptive Analytics
 OR is part of analytics: prescriptive analytics Several definitions:
 OR is the discipline of scientific methods used to make better decisions.
 OR proposes conceptual models to analyze complex situations and enables decision makers to take
efficient decision.
OR is a discipline at the intersection of mathematics, economy and computer science.  OR is naturally related to the industry and plays a vital role in its competitiveness..
© SANJAY DOMINIK JENA, ESG UQAM – MBA8419

Operations Research (OR)
 Objective:
 Propose decisions to make the best with our available means.
This assumes that we are able to:
 Take measurements
 According to certain performance indicators
 Indicators:  Costs
 Service quality
 Customer satisfaction
* De la présentation « Recherche Opérationnelle et Génie Industriel », J-C Billaut, ROADEF
© SANJAY DOMINIK JENA, ESG UQAM – MBA8419

An optimization model
(alias decision making technology)
 Objective:
 Propose decisions to make the best with our available means.
Minimize (reduce) costs
number of errors waiting time
by adjusting the decision variables production quantity of product x
frequency of bus 51 between 7h00 and 9h00 number of salespersons to hire
and respecting the constraints and requirements do not exceed the available budget
hire at least one sales person per department respect the maximum capacity of available resources
or Maximize (increase)
service frequency product offer diversity
© SANJAY DOMINIK JENA, ESG UQAM – MBA8419

Implementation of an optimization model – Who has the last word?
The manager !
 She defines the mandate of the « consultants » (internal or external)  She interacts with them to elaborate the model
 She validates the results
Analytics is an essential “tool box” of future managers
The manager needs to understand the methods and techniques used to:  Model the problem
 Solve the model
 Analyze the solution
In order to be able to
 Discuss with the consultants, understanding the important details and be able to judge whether the project was a success, or not.
 Understand which data is required to nourish the model and to make efficient decisions.
Objective of the course:
 Enable the students with the tools to play such a role.
© SANJAY DOMINIK JENA, ESG UQAM – MBA8419

Cours objectives
The objective of the course is to introduce to the student a systematic methodology to provide decision support to complex planning problems. After the course, students should know how to use such methodology in order to model some of the most common planning problems, i.e., those that are often found in practice. In this context, the course has 3 principal objectives: :
1) Understand the role of decision making technologies in an enterprise, know its possibilities and its limitations, and be able identify the circumstances in which optimization models can be useful.
2) Be able to identify and structure an optimization model given a planning context, i.e., identify the decision variables, the necessary constraints, the objective function and choose a judge the appropriateness of a solution method.
3) Have an understanding how to analyze the proposed solutions to conclude whether those are feasible in practice, and how to estimate whether such optimized solution is effectively more efficient than current planning practices.
© SANJAY DOMINIK JENA, ESG UQAM – MBA8419

Introduction of the lecturer
Associate professor, Department of Analytics, Operations and Information Technologies, ÉSG UQÀM
Experience / training::
 AXA Insurances; B.Sc. Comp Sc. (FH Köln); M.Sc. R.O.(PUC-Rio); Ph.D. R.O. (UdeM); Postdoc. (MIT SMART)
Expertise and research interests:
 Operations research; mathematical optimization
 Optimization under uncertainty
 Data science and machine learning
 Applications: logistics and transportation; facility location; revenue management; project management
Affiliation to research centers:
 Interuniversity research center for entreprise networks, logistics and transportation (CIRRELT)  Excellence research chair in Data Science for Real-time decision making
 Research center for intelligent2 management of complex systems (CRI2GS)
Industrial partners:
 BIXI, Netlift, Cascades, JDA Labs, FPInnovations, GAPSO Analytics
© SANJAY DOMINIK JENA, ESG UQAM – MBA8419

Course plan
▪Session 1 – Introduction & LP ▪ watch videos
▪ 20h00 -22h30: live course / Q & A ▪Session 2 – Network models
▪ watch videos
▪ 20h00 -22h30: live course / Q & A
▪Session 3 – Integer programming
▪ watch videos
▪ 20h30 -22h30: live course / Q & A
▪Session 4 – Revision
▪ 20h30 -22h30: live course / Q & A
▪Final exam (individual)
Evaluation
▪Final exam (individual): 100%
▪ Communication during exam not allowed ▪ Exchange of information between teams
for practical for not allowed
▪ Any type of plagiat or cheating: course failed.
© SANJAY DOMINIK JENA, ESG UQAM – PLAN DE COURS MBA 8619

程序代写 CS代考 加微信: powcoder QQ: 1823890830 Email: powcoder@163.com