Using Artificial Intelligence Principles to Strengthen Quality, Resource and Requirements Management

1) Quality management in a project environment

Quality management ensures that project outputs are fit for purpose and meet agreed requirements. It comprises quality planning, quality assurance and quality control, each serving a distinct role (Association for Project Management, 2019). Quality planning defines standards, acceptance criteria and methods, while quality assurance focuses on confidence in processes, and quality control checks actual outputs against requirements. AI can assist quality planning by analysing requirements, identifying ambiguous acceptance criteria and aligning standards across work packages. Used appropriately, this reduces the risk of defects being built into deliverables rather than detected later (PMI, 2021).

2) Quality assurance vs quality control

Quality assurance (QA) and quality control (QC) are often confused. QA is process focused and preventative, ensuring that suitable methods and controls are in place, whereas QC is output focused and detective, confirming that deliverables meet defined criteria (ISO, 2015). AI can support QA by reviewing process compliance, identifying deviations from agreed standards or highlighting weak documentation. In QC, AI can assist by checking deliverables for completeness or consistency against acceptance criteria. In both cases, AI outputs must be reviewed by competent professionals to preserve accountability and avoid over-reliance on automated checks (APM, 2019).

3) Resource management and control

Resource management involves planning, acquiring, managing and controlling people, equipment and materials required to deliver the project. Ineffective resource planning can undermine schedules and quality simultaneously (PMI, 2021). AI can support resource forecasting by analysing historical data, identifying capacity constraints and modelling alternative allocation scenarios. This helps project managers explore options earlier, improving decision quality and stakeholder communication. However, resource decisions must also consider human factors, organisational priorities and contractual constraints, which cannot be delegated entirely to automated analysis (ISO, 2018).

4) Resource smoothing vs resource levelling

Resource smoothing and resource levelling are techniques used to address resource constraints, but they differ in intent and impact. Resource smoothing adjusts activities within available float to reduce resource peaks without changing the critical path, whereas resource levelling may extend the schedule to resolve overallocations when resources are limited (PMI, 2021). AI can help analyse trade-offs by modelling scenarios and visualising impacts on time, cost and risk. Options available to the project manager include resequencing tasks, adjusting assignments, using overtime, engaging temporary resources or renegotiating priorities with sponsors (APM, 2019).

5) Requirements gathering and prioritisation

Clear requirements are essential to defining scope and quality expectations. Techniques such as MoSCoW prioritisation and T-shirt sizing support early alignment by clarifying what is essential, desirable or optional, and by providing quick estimates of relative effort (Clegg and Barker, 2019). AI can help structure requirements, detect duplication and test consistency between requirements and objectives. It can also support prioritisation discussions by grouping requirements and highlighting dependencies. Final prioritisation decisions must remain collaborative, reflecting stakeholder needs and organisational strategy (PMI, 2021).

6) Product Breakdown Structure vs Work Breakdown Structure

A Product Breakdown Structure (PBS) defines what products must be delivered, while a Work Breakdown Structure (WBS) defines the work required to deliver them (APM, 2019). Confusing the two can result in missing deliverables or poorly defined tasks. AI can support the creation of both structures by checking completeness, identifying logical grouping and testing alignment between products and activities. Learners must still develop these artefacts themselves to build competence and demonstrate understanding, using AI as a review and refinement tool rather than a shortcut.

7) What to avoid

AI should not be used to bypass stakeholder engagement, professional judgement or governance controls. Automatically accepting AI-generated quality checks, resource plans or requirements prioritisation risks embedding flawed assumptions into delivery (NIST, 2023). Outputs must be validated, documented and approved through agreed processes to maintain assurance and credibility.