Industrial Data Science and Machine Learning Research

At the end of each lecture, students will prepare a short (~2-3 pages, 12 point single space) report that addresses as many of the following questions as are relevant:
 Describe the market sector or sub-space covered in this lecture.
 What data science related skills and technologies are commonly used in this sector?
 How are data and computing related methods used in typical workflows in this sector? Illustrate with an example.
 What are the data science related challenges one might encounter in this domain?
 What do you find interesting about the nature of data science opportunities in this domain?
 Please discuss some of the characteristics of industrial data science problems alluded to in the talk and how they differ qualitatively from data science problems in other domains. (20 points of the 80 C+R points in the rubric)
Also, answer the following multiple-choice questions: You can list the quetsion number and the letter corresponding to the correct choice as Answer in your report, (2x5 = 10 pts of the 80 C+R points in the rubric)
Q1: Based on the lecture, what are the basic components of Industrial Data Science:
1. Physics/Engineering-based Models
2. Empirical, Heuristic rules & insights
3. Data-Driven techniques – machine learning, statistics, optimization, etc.
4. Cross-Dependencies and model validation
A. 1,2,3 B 1,3,4 C. 2,3,4 D. All of the choices
Q2: Based on the lecture, select all the correct statements:
1. Industrial Data Science is defined as the outcome-oriented application of mathematical & physics-based analysis & models to real-world problems in industrial operations
2. Business optimization by linking to financial & customer data is a part of Industrial Operations
3. There are many events to train data-driven models to detect anomalies and predict future events
4. User Acceptance Testing is a component of iDS framework
5. Data is often well collected as by-product of industrial systems, and readily fits the goal of enabling data-driven insights
A. 1,2,3 B. 1,2,4 C. 1,2,3,4 D. 1,3,4,5 E. All of the choices
Q3: Based on the lecture, what sort of data would be available from the asset lifecycles:
1. Design Data
2. Bill of Materials (BOM) Data
3. OEM Specifications
4. Operations Data
5. Alarms, Failures Data
A. 1,3,4,5 B. 1,2,3,4 C. 2,3,4,5 D. All of the choices
Q4. Based on the lecture, select all the correct statements about Time Series Data:
1. Time Series data is a sequence of data points, measured typically at successive points in time spaced at uniform time intervals
2. Event data is a type of Time Series data
3. The relationship to timestamp has to be maintained at all times in Time Series data
4. You CANNOT shuffle time series observations and simply pick a random sample
5. Each time series observation is NOT independent of its neighbor
A. 1,2,3,4 B. 1,3,4,5 C. 1,3,4 D. 1,3,5 E. All of the choices
Q5. Based on the lecture, there are certain aspects that can determine whether a DS project can succeed or fail, select all the correct statements from the following:
1. DS initiatives often fail if the stakeholders are ambiguous about their problems
2. DS initiatives fail if the benefits are not significant to the bottom line
3. DS initiatives fail if critical domain knowledge is not available in DS team
4. DS projects succeed if the problem has been clearly defined, and solution techniques are determined after defining the problem and exploring the data
A. 1,2,3 B. 1,2,4 C. 1,3,4 D. 2,3,4 E. All of the choices