Organizing Tasks and Roles around Machine Learning
Title : Organizing Tasks and Roles around Machine Learning
Name : Febriana Wisnuwardani
Email Address : email@example.com
Division : Organization and Management
Institutional Affiliation : University of Warwick
For decades, scholars of organization studies have put much concern on organizing work and technology (Barley, 1990; Leonardi and Barley, 2008). Many scholars also study the process of organizing work by explaining it through a more granular unit of analysis, namely tasks and roles. A number of studies have attempted to define and categorize tasks as well as theorize how tasks should be coordinated. Scholars have also attempted to study roles by considering the dynamics and interdependence between roles. Studying the dynamics of work through tasks and roles has led to useful explanations of how changes in work, especially in relation to technology, may happen.
However, in the age of algorithmic technologies, tasks are no longer only automated but rather transformed as these technologies learn from data to adapt and perform tasks. With the pervasive implementation of learning algorithms, also known as Machine Learning (ML), and the ability of ML to constantly learn to perform a new type of tasks, many scholars started to study ML implications for work. There are some tasks that can be substituted and augmented by the use of ML. As some tasks are affected, studies also show that organizations must consider the shift in role boundaries of the affected occupations. It is because many workers might leverage insights from what ML can do to re- conceptualize their purpose and identity.
Even though scholars have started to study the implication of ML on tasks and roles, hitherto, no study has explicitly examined the process of tasks and roles transformation and reorganization as the consequence of what ML achieves. Therefore, this study will examine this transformation as a process. Particularly, this study will focus on how ML analytics enlarge the reach of organizations in tracking, monitoring, deciphering, and directing the behaviours of individuals and groups and also the implication on tasks and roles within the organization.
According to a pilot study in a company, my early findings point to an interesting phenomenon. There is an indication that tasks and roles of Business Intelligence (BI) team are becoming transformed as the company implements ML. The BI team consists of Data Analysts and Data Engineers which are assigned to manage (e.g. clean, transform, categorize) data and develop rule-based algorithms to be deployed to the product. On the other hand, as a commitment to implement ML, the company hired Data Scientists which are dedicated to develop ML algorithms. The work that the Data Scientists have done in creating the ML model automates some tasks that the BI team does, such as processing and analysing data.
As a consequence, the BI team is currently starting to expand their skills in learning to build ML algorithms as well. In other words, the BI team is starting to renegotiate their roles by expanding their skills to not only convert data to insights, but also to build learning algorithms. This early finding indicates a dynamic that is happening within the BI team as they are trying to renegotiate their tasks and role boundary due to ML implementation.
This study uses a qualitative approach to understand this phenomenon better. Using participant observation, interviews, and document analysis, this study is examining the reconfiguration of tasks and roles of that particular BI team to explain how ML reorganizes tasks and roles of this team. Studying the dynamics of work in this particular team might leads to the understanding of how the process of tasks and roles reconfiguration happen due to the implementation of ML algorithms. In so doing, my research will contribute to the literature on the transformation of work (Bailey and Leonardi, 2015; Beane, 2018; Faraj et al., 2018).
Leonardi, P. M., & Barley, S. R. (2008). Materiality and Change: Challenges to Building Better Theory about Technology and Organizing. Information and Organization, 159-176.
Barley, S. R. (1990). The Alignment of Technology and Structure through Roles and Networks. Administrative Science Quarterly, 61-103.
Bailey DE, Leonardi PM (2015) Technology Choices: Why Occupations Differ in Their Embrace of New Technology (MIT Press, Boston)
Beane M (2018) Shadow learning: Building robotic surgical skill when approved means fail. Admin. Sci. Quart. 64(1):87–123.
Faraj S, Pachidi S, Sayegh K (2018) Working and organizing in the age of the learning algorithm. Inform. Organ. 28(1):62–70.