How to Approach Artificial Intelligence for Project Management
This article is part of a series of blogs on Artificial Intelligence in Project Business. See the rest of the articles in the series at the end of this blog.
Artificial intelligence (AI) has been a hot topic for many years, transforming the way businesses operate by streamlining operations, managing risks, reducing costs, and propelling innovation forward. However, AI in project-based industries, or as we call them Project Businesses, has been near non-existent. That’s not to say it’s a pipe dream or impossible. AI is achievable for Project Business if the right approach is taken.
To date, attempts at AI in Project Business (e.g. construction, engineering, ETO manufacturing, professional services) have been focused on narrow tasks, such as resource optimization or project scheduling assistance. The goal of AI in Project Business should be broader and more encompassing.
The goal should be to provide a predictive model of project intelligence that answers the question, “Will this project finish on time and on budget and to what degree will it miss or beat the plan?”
AI Takes the Guesswork Out of the Equation
Projects are filled with risks, both known and unknown, that are extremely difficult to quantify. For now, those risks are reviewed by experienced project managers who make educated guesses as to how those risks will manifest and affect the project timeline and budget.
These educated guesses of professionals today are just guesses. Their accuracy tends to vary widely depending on the person and his or her experience.
Imagine if you could use AI to replace those guesses with statistically accurate predictions. The idea is that AI would read the data from the current project and hundreds of past projects and use it to determine whether the current project will finish on time and within budget.
With this type of project-level predictions, you could completely transform risk management and forecasting. This type of AI analysis would help managers detect schedule delays and budget altering issues earlier, enabling them to take mitigating actions faster.
Humans are not good at ingesting large amounts of data and making predictions. But that’s exactly what AI does well. It can make sense of millions of datapoints in seconds, calculate the probabilities and recognize patters that humans cannot.
Current Uses of AI Across Different Industries
According to the Gartner Hype Cycle for AI, AI technologies have a long way to go until they become widespread for productive use. However, there have been some valuable applications of AI across different industries including:
Fraud Detection- to prevent credit card and identity theft.
Recommendation Engines- how Amazon knows what book you might like, and Netflix knows what new shows and movies to pitch to you.
Supply Chain Planning- how retailers replenish their stock automatically based on sales, seasonality, business cycles, current economic conditions, and even the weather.
Logistics- for optimized route scheduling and intelligent warehousing. Also, how Uber and Lyft predict when you will arrive at your destination or when your food will be delivered.
While these technologies are useful, the reason why AI cannot be applied easily in every industry has to do with how data is collected and managed. Keep in mind, AI is only as good as the data you put into it.
Structured Data is the Crux of the Issue in Project Business
To use AI to figure out if your project will finish on time and on budget requires foundational project intelligence data. Predictive AI demands massive amounts of structured data stored in a time-phased manner from which to “learn” and make predictions. Basically, the AI processes the data iteratively, looking for patterns and applying intelligent algorithms until it produces the best possible result.
Not having enough data or data in the proper format for an AI to read and make sense out of it, is useless to an AI.
While data is not lacking in Project Business, unfortunately, most Project Businesses lack a data structure that AI can use. The data is not stored and managed in a standardized way.
For AI to be able to make predictions about the future, it’s important to capture and store that data over time in precisely structured numerical formats (i.e. data that a machine can read). This is called time-series data.
Collecting and storing project management KPIs in real time is the key to creating the data foundation for predictive, project-level artificial intelligence.
Our next blog will take a deep dive into why current Project Business capabilities make it difficult to achieve AI and the requirements for successful Project Business AI.
CEO of ADEACA - Daniel founded ADEACA in 2007 to address the needs of project-driven organizations who were lacking a holistic solution for project and financial management. As one of the principal architects behind Microsoft Dynamics AX, Daniel was in a unique position to recognize the best way to fulfill this need was inside the ERP. ADEACA was born. Prior to founding ADEACA, Daniel was a principal architect and product director of Axapta at Damgaard Data, which was acquired by Microsoft in 2002 and later became Dynamics AX and now called Microsoft Dynamics 365 Operations. His experience at Damgaard equipped him with the skills to create user-centric software, an art form not practiced by the large, generic ERP companies.
In today’s increasingly complex project environments, businesses are facing unprecedented challenges in delivering projects on time, within budget, and according…
Project Business Transformation reduces risk, increases efficiency, and enhances decision-making in your project organization. Project Business Automation is a new…
Project Business Transformation is a strategic digital transformation initiative designed to fundamentally change how projects are managed and delivered in…
