.png){kind=icon}
.png){kind=icon}
Return
Optimizing digital workflow design and implementation with recommender systems for personalized efficiency
.png){kind=icon}
PRE
AI
DOI：10.1007/s10844-026-01026-x.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Organizational well-being is increasingly shaped by the evolving landscape of digital workflows, driven by a surge in innovative software applications. Among these, online automation platforms have become central to improving operational health by enabling organizations to automate business processes through flexible, task-centric workflows that minimize human intervention. Despite widespread adoption, effectively structuring these digital workflows within automation environments remains a key challenge to maintaining productivity and organizational resilience. Inspired by the success of recommendation systems in domains like e-commerce, we propose an intelligent digital workflow recommendation framework aimed at enhancing decision-making and ensuring better process alignment. Our method captures the sequential usage patterns of applications across real-world digital workflows, utilizing data from a no-code automation platform. It combines the power of singular value decomposition with machine learning-based clustering to form a robust and adaptive recommendation engine. Through extensive experiments involving various workflow recommendation stages, we benchmark our model against traditional approaches, including the Markov chain and random recommendation strategies. The results consistently demonstrate the superior accuracy and adaptability of our model across all test scenarios. Beyond empirical performance, we examine the theoretical and practical implications of our approach. By supporting better workflow design and optimization, it contributes to the broader goal of organizational well-being-enhancing operational clarity, reducing cognitive load, and fostering agile adaptation in dynamic environments. As organizations increasingly depend on AI and automation for sustainable growth and efficiency, our methodology underscores the transformative role of recommender systems in shaping intelligent, data-driven digital workflows. Integrating sequential pattern analysis with AI-powered insights, our approach becomes a strategic asset in the journey toward digital well-being and resilience.
Keywords:
Digital workflows
Recommendation
Automation
Clustering
Sequential information
Journal
J
IF:
3.4
Papers:
75
Citations:
0