Essays on Organizing Human Capital, Automation and Innovation
Ketkar, Harshvardhan Jayant
2020
Abstract
In this dissertation, I explore how different modes of organizing and automation shape processes and outcomes of knowledge-based work within organizations. With advances in artificial intelligence and automation, firms are increasingly automating knowledge-based work which were previously carried out by humans alone. The use of such cognitive tools presents implications for adopting organizations in terms of how these tools shape knowledge production, or how they affect bias and inclusion against organizational members belonging to certain groups. Correspondingly, firms are also experimenting with new forms of organizing human capital across its various projects, since deploying it judiciously becomes a source of competitive advantage. Knowledge intensive tasks that were previously carried out by experts or managers are now either being delegated to lower-level employees, or automated using algorithms, or carried out by using a combination of the two. This dissertation focuses on tradeoffs associated with new forms of organizing and automation in the context of knowledge-based tasks such as human resource allocation, integration of new knowledge, and socialization of newcomers and outsiders. I explore these questions through the three chapters of this dissertation. In chapter II, I look at how automating the integration of new knowledge affects what kind of knowledge contributions get integrated into an organization’s knowledge base. I use the context of software projects hosted on GitHub, some of which automated the process of evaluating code contributions. I find that projects that adopt automation tend to integrate narrower, component-level contributions, rather than broader, systemic contributions, perhaps because the algorithm crowds out unstructured coordination among contributors and maintainers, which is necessary for systemic contributions. In chapter III, using the same context, we look at how automation shapes inclusion and discrimination against female code contributors on GitHub. We find that after automating code review, projects tend to attract code contributions from female programmers at a greater rate than before (and compared with non-adopters). They are also more likely to eventually accept code contributions from female programmers. Finally, in chapter IV, we provide a computational model of how organization structure (open allocation or hierarchical allocation) affects allocation of human resources to available. We find that the relative balance between the organization’s human resources and the number of opportunities it faces is critical in determining the advantages of open allocation, which performs better when human resources are scarce relative to opportunities. Conversely, hierarchical allocation (with a manager) performs better when opportunities are scarce. Methodologically, I employ a combination of agent-based models and large datasets consisting of fine-grained, contribution level data to present my findings across the three chapters. Overall, this dissertation aims to better understand the impact using of different ways of organizing human capital and algorithmic automation for knowledge work within organizations.Subjects
Automation and algorithms Organization design and coordination Human Capital Evaluation and Knowledge Work Innovation
Types
Thesis
Metadata
Show full item recordCollections
Remediation of Harmful Language
The University of Michigan Library aims to describe library materials in a way that respects the people and communities who create, use, and are represented in our collections. Report harmful or offensive language in catalog records, finding aids, or elsewhere in our collections anonymously through our metadata feedback form. More information at Remediation of Harmful Language.
Accessibility
If you are unable to use this file in its current format, please select the Contact Us link and we can modify it to make it more accessible to you.