Essays in Service Operations:  Innovations and Challenges

Li, Simin

doi:https://doi.org/10.21985/n2-agfh-5z69

Work

Essays in Service Operations: Innovations and Challenges

Public

Download PDF

Download All Files (.zip)

This dissertation contains three essays that study the operational challenges and business innovations in service industries. We investigate how service providers could use wisely designed business models and service systems to induce preferable customer behaviors, and therefore manage demand and optimize revenue. Using a stylized model, we study how customers value their time, and what delay costs service providers should use to avoid suboptimal system design. Using empirical analysis, we study the operational values of an innovative online deal for the small service providers facing holiday demand swings. We investigate how a capacity expansion with limited capabilities can help manage patient flows, induce patient self-routing and thus improve emergency department efficiency. We therefore gain managerial insights in how to design a system or a business model to interact with the customers and consequently optimize performance measures. To design a service system, service providers often capture the consumers’ dislike of waiting by assuming that they incur a cost per-unit-time spent waiting. However, where does this waiting cost come from? Is it merely foregone earnings or is it something different? In Chapter 1, we address these questions by assuming an agent faces a utility maximization problem that focuses on time allocation subject to time and budget constraints. The cost of waiting is then the minimum monetary compensation required to keep the agent’s utility level constant as the agent loses time due to waiting. We show that the relationship between the cost of waiting and wages depends intricately on the agent’s compensation structure. We prove that agent’s cost of waiting is equal to her hourly wage only when the compensation structure is smooth. The equivalence between hourly wages and waiting costs breaks down for more general compensation structures such as fixed shifts or even piece-wise linear wages. Furthermore, we show that hourly wages can overestimate or underestimate the cost of waiting in the presence of such non-linear compensation structures. An agent’s hourly wage by itself gives little information about the agent’s true cost of waiting. Instead their compensation structure and utility function are crucial determinants of how they value their time. These results are robust to several modifications of our basic model. In Chapter 2, we focus on how local small service providers utilize new business tools offered by the online platforms to manage holiday demand shocks when limited by inflexible service capacity. Among the limited ways for small service providers to balance demand and supply, launching temporary consumer offers may be attractive. However, relatively little work has empirically examined whether and how such offers pay off for service providers. In this paper, using a comprehensive dataset from two leading deal platforms in China, we empirically study a new business model: the online deal. Service providers, who face predictable demand swings and capacity constraints, launch online deals for customers to prepay for service. Using a structural model, we show that online deals effectively facilitate demand-supply coordination through two levers, the discount and, more interestingly, the advance sales period. We do not a priori assume the sign of the demand swings, i.e. valleys or peaks. To our knowledge, using the advance sales period as revenue management tool has not been studied in the literature. The advance sales period and the discount serve two operational roles to achieve profit maximization – smoothing demand mean and moderating the demand variance. The latter reduces variance-related costs. Furthermore, our model estimates enable us to quantify the operational value of online deals. Via counterfactual analyses, we show that by using these two levers instead of solely a discount, 82.1% service providers see a mean profit improvement of 23.6%. The additional lever (i.e., the advance sales period) mitigates extreme discounts contributing to a boost in profits. The advance sales period also induces consumers to lock in their spending before the holiday, which is another reason for service providers to launch holiday online deals even when the holiday promises a demand increase. In Chapter 3, we move on to discuss how a clever service system design can induce desirable consumer behaviors, achieve favorable self-routing of consumers and thus improve system efficiency. This is studied in a healthcare setting, so consumers are patients. Many attempts have been made to make the emergency department (ED) use more efficient, but the effectiveness of various efforts is still unclear. One such approach is to open urgent care centers (UCCs). A UCC serves low-acuity conditions only and offers no inpatient or observation units. The benefit of having a UCC to absorb some low-acuity patients’ ED visits seems intuitively clear. However, partly due to the complications in analysis (in particular, unobserved patient preferences such as brand and location considerations), the existing literature shows surprisingly inconclusive results. In this paper, we partner with a major medical system and utilize a natural experiment setting to investigate the impact of a UCC opening on an ED. The UCC is located adjacent to the treatment ED, and both facilities are operated by the same medical system. Our data spans the opening of the UCC. We also have data for other EDs in the same medical system that are not paired with UCCs that thus serve as a control group. This allows us to use the difference-in-difference framework to tease out conclusively the impact of UCC on the ED. We find that the volume of the low-acuity patient visits to ED at the treatment hospital decreased by 13.7% after UCC opens. Low-acuity patients are effectively diverted away from the ED. Furthermore, we are the first to look at the impact of a UCC on the mix of patient. We show that the proportion of hospitalized low-acuity patient visits to the ED increases by 7.5% after UCC opens. This suggests that the patients are sophisticated enough to make a judgment on whether they would need follow up care and thus self-route to the appropriate facility to get the care they need. Our results suggest that a wisely designed service system---an ED with a co-located UCC equipped with limited capabilities---disproportionately affect the low-acuity patients who do not need to be hospitalized. Subsequently, we see an 11.3% shorter door-to-doctor time among the high-acuity patients, which suggests that costly ED resources are freed up for the more severe cases. An efficient self-selection of patients among the healthcare facilities is induced by the design (co-location, different capabilities, and different throughputs) of the service system. In turn, the service system achieves better efficiency.

Creator