IEEE Power & Energy Magazine - May/June 2018 - 70

business partner role, the utilities would be looking for those
who can convince executives, who have been doing things in
exactly the same way for 30 years, that there is a better way to
do this or that. with the addition of these business partners,
the analytics team is expected to turn capability into insights
and value-added corporate actions.

Energy Data Scientists
a data scientist is "a hybrid of data hacker, analyst, communicator, and trusted adviser," as defined by a 2012 Harvard
Business Review article. The article also recognized that "the
combination is extremely powerful and rare," which is further confirmed by the newVantage partners big data executive survey 2012 among fortune 1000 c-suite executives
and federal government leaders. while 70% of them were
hiring or planned to hire data scientists in the near future,
only 2% of them reported that they had no challenges finding the right big data talent and skilled resources.
in october 2017, kaggle released the results from its 2017
survey of data science and Machine Learning. Table 1 provides some statistics about the data scientists in the United
states and around the globe. simply speaking, a typical data
scientist in the United states is around 32 years old, has a master's degree, and makes about Us$110,000 a year.
Today, there are more than 175 programs in the United
states offering graduate degrees in analytics and data science. some of them are quite transparent with their student
profiles and employment data. for instance, in the 2017 graduation class of the master of science in analytics program
(Msa) at north carolina state University, 107 students were
placed by graduation with a median annual base salary of
Us$91,000 plus a median signing bonus of Us$7,500. The
median age of the class was 25.
one consensus between the kaggle survey and north
carolina state University's Msa employment report is obvious: data scientists are getting high salaries! The pay checks
are also reflecting the supply-demand relationship in today's
job market. The data scientists are in very high demand. given
the talent gap identified in the 2012 big data executive survey,
it should be no surprise that they are getting well paid in the
early stage of their career.

table 1. Data scientist statistics, according to a
2017 Survey of Data Science and Machine Learning.

United
States

World

Median age

Median annual salary

$110,000

$55,441

Highest level
of formal
education

Bachelor's degree

26.5%

32%

Master's degree

44.5%

41.8%

Doctoral degree

20.7%

15.6%

ieee power & energy magazine

on the other hand, utilities have been struggling with
the electricity workforce for more than a decade. according
to the Qer, the age distribution in electric and natural gas
utilities in 2014 is bimodal, with the lower peak in the age
group 33-37 and the higher peak in the age group 53-57. The
median age is in the 40s, while the average annual income
of the electricity sector in 2015 was $106,000. a power engineer in his 40s would earn a similar salary as a typical U.s.
data scientist at age 32 or a new graduate from north carolina state's Msa program in his/her late 20s.
since most of the data scientists today are in the business
sectors that adopted analytics early, how can utilities compete on talent acquisition? apparently those world-class analytical talents know to go after silicon Valley and wall street
for high-paying jobs. Most of them probably never thought
about working at a regulated utility, commonly known as
a slow-moving monopoly. one strategy currently adopted by
several utilities is to proactively go hunting for these data
scientists. They go out and search the best possible graduate
programs in universities for specific subject areas. among the
group of ph.d.s working in these programs, they determine
the ones they want to target and then get in touch with them
directly and get their attention.

State of Power Engineering Curriculum
ideally, university power programs are supposed to provide fresh graduates with skills that meet the current and
emerging needs of the power industry. however, the mainstream power engineering curriculum in the past has rarely
recognized analytics as a crucial component. This is partly
reflected by the literature in power engineering education.
for instance, among the 30+ papers collected by the two
education-focused special sections of IEEE Transactions
on Power Systems, none of them had the word "analytics"
in the title. The traditional power engineering programs,
which just experienced three decades of decline in student
enrollments and faculty count, did not have resources to
fully accommodate advanced analytics in the curriculum.
we have surveyed the power engineering curricula in the
United states (figure 3). at the undergraduate level, many
electrical engineering programs have the required courses
that lead to a b.s.e.e. degree with a power and energy concentration during junior and senior years, such as energy
conversion, introductory power and energy systems, and
controls. Typical elective courses include green electric
energy, electric machinery, power electronics, power system
analysis, power system operation and control, power system
economics, and power system protection. while most, if not
all, of these programs list a probability and statistics course
in the b.s.e.e. degree requirements, we did not find any
courses related to energy data analytics in the undergraduate
power programs.
at the graduate level, most electrical engineering programs with a power system concentration emphasize aspects
of analysis, operations, and control, offering advanced courses
may/june 2018

Table of Contents for the Digital Edition of IEEE Power & Energy Magazine - May/June 2018