Introduction

We are researching how to employ Large Language Model (LLM) tools such as ChatGPT to create project planning documents. In Getting Real About Using AI To Support Project Management: Designing a Baseline Demonstration Project we suggested a two-phased approach for a demonstration project:

Phase One will be devoted to the creation of project planning documents with the support of an LLM tool such as ChatGPT.

Phase 2 will be a test of using a LLM such as ChatGPT to support the actual management of the project.

In this article we provide more detail about Phase One based on work we’ve done to partially automate the production of two key project planning documents:

The Business Case Document outlines management’s business reasons for applying resources to a project.

The Project Definition Document, which the LLM tool creates from the Business Case Document, outlines the steps that project managers must take to actually perform the project.

The sequence of ChatGPT-supported actions is straightforward and fast. In response to a specific problem statement, ChatGPT generates a business case based on a template the LLM helps to develop or based on a model already used by the organization. Then in response to the LLM-generated generated business case, ChatGPT generates the project definition document which can then be used for detailed planning.

By now it is well known that it is possible to interact with a tool such as ChatGPT in plain English and receive output in plain English as well. From the outset of our research, however, we have been investigating (especially Michael with his Python scripts!) how best to interact with ChatGPT while ensuring that what ChatGPT delivers responds to the actual organizational and technical context in which the planned project will be managed.

Workflow and System Components

The workflow we’re considering has three main components: the Problem Statement, the Business Case Document, and the Project Definition Document.

1. Problem Statement

The manager-supplied problem statement tells the LLM what type of project we need to create a business case for. Here are three examples, each of which can be stated in varying levels of detail:

“Create a business case for a project to implement Microsoft Project Server in my company.”

“Create a business case for a project to implement a zero-trust security model throughout our organization.”
“Create a business case to combine our organization’s three data centers into a single data center.”

When we submit such statements in raw form to ChatGPT without any special contextual information, the LLM can respond with reasonable sounding output—the draft of a business case--but potentially at a level of generality that may miss important considerations for the organization making the request. We have therefor developed a template-based approach for creating the Business Case Document.

2. Business Case Document

The business case template is designed to be processed by the LLM tool to generate a business case document that reflects the specific needs of the organization. The business case template has two parts:

An outline of the desired contents of the business case stated in terms of headers, sub headers, and paragraph (see Appendix A1 for an example).
A set of Python scripts linked to the outline that prompts the LLM tool on what each section and subsection of the business case should contain (see Appendix A2 for example).

Appendix A3 shows a sample from an output business case document.

Some organizations have specific business case models they use for evaluating distinct types of projects. These can be used as models or templates for how we want the LLM tool to respond to the problem statement. Others leave business case methodology up to contractors or consultants. Or we can use the LLM tool itself to help construct the basic business case model. We have experimented with all three approaches and have found that ChatGPT can help compare, edit, and synthesize business case approaches very quickly.

3. Project Definition Document

What we next want the LLM tto do is to read the business case and, in response, draft the project definition document. Again, much of this can be done in plain English. As with the business case document we have also found it desirable to employ a content template and a linked set of Python scripts to prompt the LLM tool how to process the business case in its generation of the project definition document. This template reflects the specific needs of the organization via two main parts:

An outline of the desired contents of the project definition document stated in terms of headers, subleaders, and paragraphs (see Appendix B1 for an example).

A set of Python scripts that prompts the LLM on what each section and subsection of the project definition document should contain, given the input from the business case (see Appendix B2 for example).

Appendix B3 shows a sample from an output project definition document.

Discussion

Each of the authors has been involved with project management and consulting for many years. Based on our experiments we believe the processes described above can significantly accelerate project planning and project initiation. Questions we want to investigate further include (1) how much faster? and (2) at what cost?

The methods described above can also be applied to a variety of other project planning documents related to risk management, issue resolution, cybersecurity, change management, communications, and other project management functions. These can all use input from the problem statement, business case, and project definition as the basis for detailed—and accelerated--planning.

We expect that much of the step by step work we have done to test out the above workflow will be eventually automated, including the automated development of prompting scripts to drive both the structure and content of output documents.

While we have been impressed with the quality of output that can be generated quickly just given the input of an English language problem statement, we are also finding it essential to incorporate human editing and review processes to capture needed corrections or changes. We are currently examining ways to incorporate editing and review processes. Given that document production can proceed so rapidly using the process outlined above, traditional delays between document drafting, document review, and document approval can, we now believe, be significantly shortened.

The testing we have done so far has focused on the creation of formatted and complete Word documents suitable for human review. Targeting online or web-based formatting instead of using the above approach should be straightforward.

Conclusion

Working faster and smarter is the goal. We believe that LLM tools such as ChatGPT will be “force multipliers” when properly used by savvy project managers. We also believe that the above process must be guided by experienced professionals and project managers who understand the needs of the organization and can express those needs in how templates and documents are constructed, edited, and reviewed.

Appendix A1: Business Case Table of Contents

Appendix A2: Business Case Scripts Sample

Appendix A3: Business Case Output Sample

Appendix B1: Project Definition Document Table of Contents

Appendix B2: Project Definition Document Script Sample

Appendix B3: Project Definition Document Output Sample

About the Authors

*Dennis D. McDonald, Ph.D.

Dennis’ writing, editing, and research work combines analytical and communication skills with extensive project management experience. His industry work includes support for civilian and military contractors, higher ed data governance, corporate database and software consolidation and retirement, adoption of collaborative technologies for knowledge sharing, and open data system development. His writing and content development work includes proposals (business and technical), project planning, research reports and white papers, and market research. He is a member of Alexandria Virginia’s Public Records Advisory Commission and a volunteer for the Alexandria Film Festival. He has served as a private school Board of Trustees member and has been a part owner of an IT-focused consulting company. He is a AAAS member and CMMC Registered Practitioner (RP). His web site is located at www.ddmcd.com and his email address is ddmcd@ddmcd.com.

**Michael Kaplan, PMP

Michael Kaplan is a highly successful Senior Program Manager with over two decades of experience delivering impressive results for enterprises. His expertise in program management, project management, and digital transformation has enabled him to protect the national power grid and secure the critical systems of six global banks against cyberattacks. He has also played a pivotal role in enabling 30 organizations to transition into digital enterprises, implementing VMware software and leveraging Agile, Waterfall, and hybrid methodologies. Michael has excelled in team management, leadership, communication, and collaboration. He is certified as a Project Management Professional, Certified Scrum Master, and ITIL, and proficient in the use of various tools. Michael has worked with a broad range of clients in the financial and insurance sectors, healthcare, educational institutions, and government agencies. His natural talent for connecting the dots across people, process, and technology domains, his commitment to delivering excellence, and his professionalism make him a highly respected Program Project Management Consultant. He can be reached via email at kaplan.usa@gmail.com.