Pricing and Job Allocation in Online Labor Platforms

Abstract

Problem Definition: We model, analyze, and optimize the operations of an online labor platform that matches jobs to workers. The arrival of jobs and workers to the platform is stochastic and the job processing time is random. The platform chooses the fees per job and assigns jobs to workers with the goal of (i) maximizing platform revenues, (ii) minimizing the unpredictability in workers' profits, and (iii) minimizing any delay in processing the incoming jobs. Workers are sensitive to the revenue they make in the platform and, therefore, the worker arrival rate depends on the platform's pricing and job allocation strategy. Academic/Practical Relevance: We contribute to the fields of online platform operations and stochastic systems. The asymptotically optimal policy that we introduce is simple and practical. Our analysis provides new insights into the management of uncertainty in online platforms and how uncertainty impacts objectives (i), (ii), and (iii) above. Methodology: We propose a continuous-time stochastic model that describes the mechanics of the platform. We introduce a policy that sequentially allocates jobs to workers in a rotating manner. The optimal parameters of this policy are corrected versions of the fluid solution. Then, through an asymptotic analysis, we prove that this policy is optimal as the platform scales. Finally, we examine the policy's performance through a discrete event simulation. Results: We introduce a policy class called Uniform Allocation (UA) and provide an analytical characterization of the platform's behavior and performance under this policy class. Then, we design a UA policy that simultaneously optimizes objectives (i), (ii), and (iii) as the system scales. We obtain insights into the UA policy's behavior through a discrete event simulation and find that it leads to similar profits but much lower worker income variability compared to other policies. Implications: Our proposed pricing and assignment policy align objectives (i), (ii), and (iii) above. From an academic standpoint, we demonstrate how the platform's revenue maximizing pricing and allocation policies can also act as a form of risk pooling. From a practical standpoint, we determine that platform revenue maximization is not incompatible with predictable profits and stable schedules for workers.