Editorial Statement

The INFORMS Journal on Data Science (IJDS) is a peer-reviewed journal, aiming to publish top innovative and potentially impactful data science methodologies contributing to decision making in engineering, business, management, and industry.

By curating and publishing state-of-the-art generalizable knowledge, IJDS aims to provide a dedicated focal point for important data science research in socio-technical aspects of engineering, business, management, and industry, to benefit the scientific community, industry, and society at large. IJDS strives to elevate the visibility of INFORMS data science communities.

Data Science Methodologies

The INFORMS data science research community in business schools, industrial engineering departments, and in industry research groups has led to many important methodological contributions to management and engineering research and applications. IJDS aims to publish such top research, which uses a range of data science methodologies including statistics, machine learning (ML), operations research (OR), engineering, econometrics, and other computational disciplines, such as physics. Causal, predictive, descriptive, and prescriptive methodologies are all of interest. Interdisciplinary contributions are especially encouraged.

Data science research for decision-making environments is motivated by decision-making challenges and the availability of new types of data. IJDS welcomes manuscripts on research motivated by a relevant, real-world decision making challenge, and that introduces a novel data science methodology or approach where data is focal. This combination distinguishes IJDS papers from those suitable for data science journals in other disciplines as well as from domain-based journals.

We are witnessing an explosion of social, economic, legal, environmental, and ethical implications associated with the adoption of data science methods by industry and government, utilizing the enormous amounts of now-available micro-level behavioral and industrial data (big data). Algorithmic decision making has now expanded from direct marketing and credit scoring to new contexts with high social and managerial stakes such as hiring and retention, judicial decisions, and political persuasion. New industrial and engineering applications such as smart cities, smart transportation, personalized health, Industry 4.0, and digital services have introduced new types of data, such as multi-stream, multi-resolution data of spatio-temporal characteristics. These call for development of domain knowledge-guided methodologies to serve the full data-to-decision process, often in real time.

IJDS aims to publish top quality research in these domains. An IJDS submission should therefore include four key components:

Data: real-world or simulated
Models/algorithms: innovative data science methodology (model/algorithm/approach)
Managerial/engineering/industrial relevance: decision-making motivation and potential/actual impact
Implications: consideration of relevant practical (e.g., computation, implementation) and ethical (e.g., societal, environmental) implications

Please note that the first two key components must be explicitly discussed in all regular research paper submissions. The 3rd and 4th components may be implied, for example, when a paper addresses technical aspects of a well-established machine learning or data science method with well-documented applications in engineering or business.

Conference Proceedings and Working Papers

IJDS will consider for publication research that was presented or published in peer-reviewed conferences and was then further developed into a journal manuscript. The manuscript must include a valuable contribution beyond that in the conference paper, and authors must share the conference paper and a clear description of how the submitted manuscript differs from the conference paper.

Data Science Ethics and Reproducibility

With data ethics becoming critical in both industry and academic research, IJDS intends to play a key role in educating and disseminating ethical data science practices. Toward this end, guidelines on reproducibility and data use ethics will be inherent to the IJDS pipeline. Authors of accepted papers should share any datasets and code used for generating the results reported in the paper and complete a reproducibility workflow to facilitate the reproduction of the research results by potential users. In the case of confidential data, alternative reproducibility plans may be acceptable, subject to approval at the discretion of the Editor-in-Chief.

Editorial Process

To facilitate timely publication of state-of-the-art research, the IJDS review process is designed to provide a rigorous, fair and timely review process. The journal provides clear, publicly available review guidelines and criteria assessed, aiming to reduce authors’ uncertainty, optimize reviewers’ efforts, reduce revision cycles, as well as guide reviewers and the editorial board on the type of feedback that can lead to a focused, fair, and efficient reviewing process. All papers submitted to the journal are first assessed for editorial fit with the IJDS mission and standards. Papers that are aligned with the journal’s goals and standards are assigned to a Senior Editor, who is responsible for managing the review process. The Senior Editor works with an Associate Editor and reviewers, who provide feedback to authors. IJDS strives to issue the first decision to 90% of the submissions within 60 days, and a final determination in no more than two rounds of revision, although exceptions may occur in rare instances.

Quicklinks & Resources

Reviewer Guidelines

Available Issues