Content uploaded by Robert Prieto
Author content
All content in this area was uploaded by Robert Prieto on Sep 18, 2020
Content may be subject to copyright.
Content uploaded by Robert Prieto
Author content
All content in this area was uploaded by Robert Prieto on Apr 27, 2020
Content may be subject to copyright.
Proper Reliance on Articial
Intelligence in Project
Management
Key Points
!"
!
!!"
Introduction
!
"!"#$%&
'""!(
!
))"!"!
!
(!!!
$%&*!
+"!
!,!"!
"
)!
1 NAC Executive Insight Impacts of Artificial Intelligence on Management of Large Complex Projects.
2 NAC Executive Insight Artificial Intelligence Ethics in the Project Management and Civil Engineering
Domains
!
-''"*.'/
AI and Project Management
0!1!
"
!(,,,
!"
"2!3"
""4
5(
!6
-Strong AI predictive tools, with known confidence levels at various time frames (Less
confident prediction of failure early on, but with a strengthening predictive confidence as
more time lapses), including:
-Transparent and robust AI algorithms, trained on known,
relevant data sets and validated for intended use.
-Knowledgeable deployment of validated AI to use-cases,
verified to be consistent with the validated AI.
-Recognition of AI limitations due to excluded data
(external ecosystem data) and an assessment of the relevance of
its consideration in the particular use-case (project).
-Effective and meaningful project reviews, undertaken regularly and using AI
predictions to focus and strengthen the depth of project reviews and diagnosis. AI tells us
the project has a “fever,” but management, especially more senior levels of management
with broader and more holistic views, must seek the underlying causes and develop a
treatment plan. An NAC Executive Insight3 on project reviews discusses the adverse
impacts from perfunctory or nonexistent project reviews, which are becoming all too
common.
-Decisive action, not delayed by a defensive response to what the AI is saying, but driven
by the earliest diagnosis obtained from the now heightened regular project review process
and any “deep-dive” review it may trigger. AI will produce “false positives,” or maybe in
this context the term “false negatives” may be more apt. Its ability to predict project
success is not yet well established, which may speak more to the nature of large complex
projects than artificial intelligence. Even when the AI has made the right call, raising
3 NAC Executive Insight Effective Project Review Meetings
!
-''"*.'/
concerns on project performance and trajectory, initial diagnosis and treatment plans may
evolve as more insight becomes available. The value of time, afforded by the AI’s
predictive analytics, must not be lost.
-Independent audit
))
"
"!
7)!"""
"
About the Author
85$!%*.''9
1!
""
!
-''"*.'/
Although the author and NAC have made every eort to ensure accuracy and completeness of the
advice or information presented within, NAC and the author assume no responsibility for any
errors, inaccuracies, omissions or inconsistencies it may contain, or for any results obtained from the
use of this information. The information is provided on an “as is” basis with no guarantees of
completeness, accuracy, usefulness or timeliness, and without any warranties of any kind whatsoever,
express or implied. Reliance on any information provided by NAC or the author is solely at your own
risk.
!
-''"*.'/