Robust Dynamic Pricing for Multiple Perishable Products

Description

Chronic illnesses such as cancer and diabetes are leading causes of death globally, resulting in substantial economic costs for many countries. Smart health devices and applications (SHDA) and technological advances in artificial intelligence (AI) are considered promising approaches to reducing healthcare costs and improving chronic illness care and health communication. AI-powered SHDA is being used as a surrogate for formal support that medical professionals typically provide. Through deep analysis of extensive patient data, SHDA can formulate and deliver personalized care messages to patients in apps that family members can also use for informal support of these patients. Few studies, however, have investigated whether or how SHDA affect health behaviors and psychological wellbeing, two key outcomes of illness management.To fill this void, this project extends optimal match theory of social support and matches the three main aspects, type of support, source of support, and stage of patients’ illness management with patient needs. Working with a start-up company operating an AIpowered SHDA platform for diabetes management in mainland China, participants will be drawn from about one million diabetic patients who use the platform daily. Study 1 will use a multi-wave survey to measure how the two types and two sources of support messages delivered via SHDA apps can jointly influence patients at the early and late stages of illness management. Study 2 will conduct a field experiment to validate Study 1 results, comparing the types (informational/ emotional) and sources (app/ family member) of patient support at these two stages. In another field experiment, Study 3 will explore whether AI can effectively act as a surrogate or supplementary tool for family members in delivering emotional support to patients. Four versions of these support messages will be designed and delivered via the company’s SHDA platform. Participants’ health behaviors and psychological states will be measured several times over one year.To the best of our knowledge, the project will be the first to develop a theory-based contingency model for AI-powered SHDA and empirically examine the matching effects of the social support types and sources. Results will provide managerially actionable guidelines on how social supports for patients can deliver optimum chronic illness management. By uniting patients, healthcare providers, and family members to combat and manage patients’ chronic disease, the divide between informational and emotional supports provided by AI-powered SHDA apps and the (patient-preferred) “human touch” offered by family members and close others can be narrowed.