Artificial Intelligence research

This year we presented at one conference and published three articles on artificial intelligence (AI) projects: success factors, stakeholders, and accountability. We considered the impact of the actions of project participants on society and the environment. In summary, project sponsors, project managers, & team members implementing AI systems are moral agents accountable for the harms (minor to serious, intentional or unintentional) or benefits of the systems they develop.

AI projects need:

  • Clear scope
  • Passive stakeholder representation
  • Specialized, diverse team
  • Ethic practices
  • Systematic record-keeping

Should avoid:

  • Moral hazards with suppliers
  • Black box designs
  • Moral buffers in decision-making

Must protect:

  • Data & Privacy
  • Financial & legal interests
  • Human & civil rights
  • The environment

You can find the open-source version of the articles here:

Stakeholder-Accountability Model for Artificial Intelligence Projects

Stakeholder Roles in Artificial Intelligence Projects

Artificial Intelligence Project Success Factors—Beyond the Ethical Principles

The paper Artificial Intelligence Project Success-Emerging Trends was presented at TEMSCON 2022: Societal Challenges: Technology, transitions and resilience virtual conference, April 25 – 29.2022, Izmir, Turkey.

Agile & Project Management Survey Results

The Purpose

The success rate for agile methodologies is on par with, if not better than, those managed under a traditional methodology. In addition, enterprise agile frameworks are at the peak of adoption. Thus, if agile methodologies are followed rigorously and exclude a project manager, then maybe the project manager role and some project management tasks are obsolete. The aim of the research was to answer the following questions:

  • Are project managers engaged in agile projects?
  • Who executes the project management tasks in projects applying agile methodologies?

The survey report summarizes the survey inputs from and analysis of 120 projects. The first section provides descriptive statistics for the data that was collected as part of the survey. The second section provides summary of the analysis that was performed with the survey data.

The Participants

The participants were from 20 industries with no geographic region having an overwhelming majority. The majority (81%) of the projects were undertaken within the last five years and lasted more than one year (56%). Most of the projects (81%) had less than 21 team members.

The Results

Scrum and waterfall were the top methodologies at 22% and 20%, respectively. However, the different types of agile methodologies (a single methodology, multiple agile, or a scaled agile framework) represented 46% of the cases. There was no significant difference in time, cost, requirement, or overall delivery performance between the agile and non-agile projects.

The project manager role was involved in 67% of the projects, including 58% of the agile projects, 82% of the mixed methodology projects, and 79% of the plan-driven projects. The agile coach, product owner, and team combination – a full scrum team – was present in 23% of the projects. However, the combination of roles could be found in almost all methodologies, except kanban and other plan-driven methodologies.

Based upon a mapping of the standard project management processes to the principles from the agile manifesto and the scrum roles, a consolidated view of project management responsibilities for scrum projects was created. In some cases, it is the practice of the method itself that is responsible for realizing the activity, while in other cases, it is a specific project role.

The Bottom Line

Project management remains an important and significant set of activities in agile and non-agile projects. The team, product owner, and project sponsors are taking on the informal role of some project management tasks. The agile coach is not a substitute for the project manager. Yes, project managers are engaged in agile projects.

Project Stories

Share your project story to add to the conversation and help explain why projects succeed or fail:

Benchmark Report: BI, BigData, & Analytic Project Success

We have created a template for a benchmark comparison report based upon the survey input. This report compares a benchmark project to the summary of the survey inputs and analysis. The survey analyzed 78 projects for critical success criteria and factors and created a classification model. Based on the model, the classification of the project is given and comparisons are made to other projects in the survey. 

Figure 1: Sample Benchmark Report

Decision Support Projects Benchmark Report6413987601

The classification model has been documented and peer reviewed by members of the Computer Science and Information Systems community.  In this form, the report is only available to people who participated in the study.




Survey results: BI, BigData, & Analytic Project Success

The aim of the research was to understand the success criteria for decision support projects and what influences the performance of those projects. “Decision support projects are implementation projects that deliver data, analytical models, analytical competence, or all three, for unstructured decision-making and problem-solving. They include subspecialties such as big data, advanced analytics, business intelligence, or artificial intelligence” (Miller,2018). This report summarizes the survey inputs from and analysis from 78 projects. The first section provides descriptive statistics for the data that was collected as part of the survey. The second section provides a summary of the analysis that was performed with the survey data.


The majority of the projects were undertaken as internal projects by large organizations, with big teams and networks of involved organizations. They were diverse in terms of complexity, pace, novelty, and team structure. The participants were from 22 countries with 73% being based in Europe.

Project Classifications

Analytic competency and building analytical models and algorithms are characteristics that differentiate the decision support project types.

Critical Success Factors

System quality and information quality are critical success factors that influence system usage and system usage influences project success. Project schedule and budget performance are not correlated with the other success measures so they are not critical success factors in most cases.

Stakeholder Contribution

Business user, senior manager, top management, and data scientist participation in project activities such as requirements and model building is a benefit. It increases the chances of achieving organizational benefits months or years after the project has been completed.


The recommendation is to actively engage business users and senior managers in hands-on project work such as building models and to focus on providing sufficient system and information quality.  As a consequent, the project should deliver long term organizational benefits.

Next Steps

On the following dates, we will discuss the study results in our office in Heidelberg, Germany. Please contact us if you wish to join.

  • Success Factors for Business Intelligence (BI) and Big Data Projects: Friday, 28.09.2018
  • Impact of Multidisciplinary Teams and Data Scientists on Project Success: Friday, 26.10.2018
  • Success Criteria for BI and Big Data Projects: Friday, 30.11.2018
  • Stakeholder Influence on System Use and Success for BI and Big Data Projects: Friday, 25.01.2019