On-demand Service Platform Research Articles

Reward and Consumer Brand Engagement: Building Brand Loyalty in The On-Demand Service Platform

Reward and Consumer Brand Engagement: Building Brand Loyalty in The On-Demand Service Platform

Customer and provider bounded rationality in on-demand service platforms

The growing literature on operations management in the context of the sharing economy typically assumes that both customers and providers are fully rational. In contrast, we consider an on-demand service platform (e.g., Didi and Uber) with boundedly rational customers and providers that sets a price charged to customers and a wage paid to providers. Both customers and providers are sensitive to the payment terms set by the platform and also to congestion in the system (given by the relative numbers of available customers and providers in the market). We capture bounded rationality using a model in which customers and providers are incapable of accurately estimating the congestion level. We examine the impact of bounded rationality on the platform profit, consumer surplus, and labor welfare. We find that both customers’ and providers’ bounded rationalities may benefit the platform. Specifically, when customers’ or providers’ bounded rationality level and service valuation are relatively large or the valuation is relatively small, more irrational customers or providers increases the platform’s profit. Moreover, we find that the platform can exploit the bounded rationality differences between customers and providers to gain profit. Counterintuitively, we also demonstrate that the high bounded rationality of customers or providers may increase consumer surplus and/or labor welfare. Finally, bounded rationality on one side (e.g., customer side) can make bounded rationality on the other side (e.g., provider side) more likely to increase the platform’s profit, consumer surplus, or labor welfare under certain conditions.

Fast Forecasting of Unstable Data Streams for On-Demand Service Platforms

Fast Forecasting of Unstable Data Streams for On-Demand Service Platforms

Deterministic and probabilistic service-hailing mode choice for on-demand service platforms

To cater to the diverse needs of consumers, on-demand service platforms often provide vertically differentiated services and allow consumers to choose their preferred service class. In the traditional service-hailing mode, consumers choose from vertically differentiated services in a deterministic way. However, an emerging service mode bundles options to allow consumers to choose multiple classes of services simultaneously, after which the system assigns one class, which seems probabilistic from the consumer’s point of view. This paper elucidates the differences between the deterministic (DE) and probabilistic (PR) service-hailing modes in terms of the decision-making and profitability of the on-demand platform. It also investigates the platform’s optimal pricing decisions under the DE and PR modes and service mode choices. This paper constructs a stylized model capturing the participation and interaction behaviors of consumers and self-employed service agents, assuming endogenous supply and demand. The model is used to characterize the platform’s optimal price and wage decisions for vertically differentiated services. The results show that the platform’s pricing strategies and service targets are jointly determined by demand-side and supply-side factors, and whether the platform should adopt DE or PR mode is contingent on the consumers’ basic valuation level and its volatility. Furthermore, exploring a more complex hybrid mode suggests that introducing the bundled option does not necessarily boost the platform’s profit. Our findings offer meaningful insights into the pricing strategies and service-hailing mode selections of on-demand platforms offering vertically differentiated services.

The effect of gamification elements on user buying intention in an on-demand service platform

Gamification increases user interaction in many industries, including on-demand services. Gamification's effects on on-demand service platforms need further study. This study explores gamification elements' influence on user buying intention on an on- demand service platform. This research extends the theoretical model of the Technology Acceptance Model by incorporating gamification elements, satisfaction, and engagement. Data were collected by survey and analyzed with SEM GSCA. This study found that points and rewards positively and significantly indirectly impact user buying intention. This study expands gamification research to the on-demand service platform sector, needing more scientific research.

Behavior-Based Pricing in On-Demand Service Platforms With Network Effects

Many on-demand service platforms have begun to implement behavior-based pricing (BBP). Unlike traditional firms, these platforms have network effects and function as intermediaries that connect providers with consumers, while not charging any fees to providers. In this study, we consider two horizontally differentiated on-demand service platforms and examine how BBP adoption affects the platforms’ profit and consumer/provider surplus under network effects. The results show that network effects and the discount factor play critical roles in determining whether BBP can yield higher profits. When the platforms are sufficiently myopic, and the strength of cross-side network effects is relatively weak, using BBP generates more profits; otherwise, using BBP makes the platform worse off. BBP always creates higher consumer surplus, reduces the average consumer surplus of switchers, and increases the average consumer surplus of repeat consumers. BBP always decreases the surplus of providers serving switchers and conditionally increases the surplus of providers serving repeat consumers. Under asymmetric BBP, platforms will not always engage in poaching the rival's consumers and consumers are worse off with poor services. Using asymmetric BBP can be lose–lose, lose–win, or win–win for platforms and consumers, which depends on the relationship between the perceived value, network effects, and service quality.

Set a Goal for Yourself? A Model and Field Experiment With Gig Workers

On-demand service platforms are interested in having gig workers use self-set, nonbinding performance goals to improve efforts and performance. To examine the effects of such self-set goal mechanisms, we build a behavioral model, derive theoretical results and testable hypotheses, and conduct a field experiment using a large gig platform for food delivery. Our model analysis finds that individual workers’ optimal self-set goals may exhibit a spectrum of difficulty levels, ranging from trivial to impossible, depending on workers’ reference-dependent utility coefficients and self-control cost. Moreover, workers’ efforts are higher with properly set goals rather than no-goals. Consistently, our experimental data show significant treatment effects of self-goal setting, and a causal tree algorithm identifies subgroups who are mostly motivated by self-set goals. Furthermore, our study compares two common types of performance metrics for goal setting: the number of completed orders and total revenue. Our model suggests different cases of effort and performance improvement for the two goal types. The experimental data suggests that both goal types improve efforts equally but lead to different attainment rates. Specifically, the goal attainment rate is lower for the revenue-goal treatment than for the order-quantity-goal treatment. Further analysis reveals that this disparity is due to workers setting excessively high revenue goals. Our study demonstrates the efficacy and limitations of self-goal-setting mechanisms and yields two important managerial implications. First, the implementation of self-goal-setting mechanisms could improve gig workers’ efforts and performance. Second, encouraging order-quantity goals instead of revenue goals could help gig workers achieve higher attainment rates.

Managing priorities on on-demand service platforms with waiting time differentiation

Inspired by recent applications, this article considers the problem of an on-demand service platform (e.g., Uber, Lyft, Deliveroo) that differentiates the service of customers according to their willingness to wait and to pay. Upon arrival, customers can be admitted to the system or rejected and, at any time, service providers can be allocated to admitted customers or reserved for future customers. We formulate this problem as a continuous-time Markov decision process, which is used to find the optimal policy. Exploiting properties of the value function, we first show that the optimal policy is described by state-dependent admission thresholds and a strict priority rule with respect to service provider allocation. In a numerical study, we then compare the performance of the optimal policy with simpler policies. We find that the platform, the customers, and the service providers can benefit from a policy that controls customer admission and service provider reservation. Finally, some evidence of the robustness of the policy is provided in two extensions.

Exploring Deep Learning Approaches for Short-Term Passenger Demand Prediction

An accurate short-term passenger demand forecast makes a contribution to the coordination of traffic supply and demand. Forecasting the short-term passenger demand for the on-demand transportation service platform is of utmost significance since it might incentivize empty cars to relocate from over-supply regions to over-demand regions. Yet, because spatial, temporal, and exogenous dependencies need to be evaluated concurrently, short-term passenger demand forecasting may be rather difficult. This article aims to investigate several methods that can be utilized to forecast short-term traffic demand, with a primary emphasis on deep learning approaches. We examine varying degrees of temporal aggregation and how these levels affect various architectural configurations. In addition, by analyzing 22 models representing 5 distinct architectural configurations, we illustrate the influence of varying layer configurations within each architecture. The findings indicate that the long-term short memory (LSTM) structures perform the best for short-term time series forecasting, but more complex architectures do not significantly enhance the outcomes. Moreover, considering the spatiotemporal aspects results in an improvement in the prediction of more than fifty percent. In addition, we investigate the vectorization of time, also known as Time2Vec, as a way of embedding to make it possible for a selected algorithm to recognize periodic characteristics in time series, and we show that the outcome is improved by fifteen percent.

Strategic driver’s acceptance-or-rejection behavior and cognitive hierarchy in on-demand platforms

It is critical for an on-demand platform to understand its users’ strategic-decision behaviors and thereby design a good policy to match the supply with the demand. Through behavioral game-theoretical modeling, we investigate the effects of drivers’ strategic acceptance-or-rejection behavior on their incomes and the overall platform users’ welfare, in an on-demand service platform with two service locations (i.e., a hot area and a cold area), wherein the competition between drivers arises due to overall supply exceeding overall demand and the location uncertainty of drivers. We first construct a full-rationality model to investigate drivers’ decision strategy in a competitive environment and attain the Nash Equilibrium outcomes in the stationary state of the system in the long run. We then employ the Cognitive Hierarchy theory to capture the heterogeneity effect in drivers’ strategic acceptance-or-rejection behaviors, which is more realistic than the full-rationality model. We find that the predictions of the Cognitive Hierarchy model can approach the Nash Equilibrium outcomes under an extreme condition. We can conclude that (i) the optimal acceptance probability of the drivers does not decrease with the rejection penalty or with the relative popularity of the cold area under our research settings. (ii) From the profit/welfare perspective, increasing the popularity of the cold area is beneficial for the platform users; however, a higher rejection penalty may not be always better for the platform users as a whole. Managerial implications are discussed.

The Dual Effects of Team Contest Design on On-Demand Service Work Schedules

Emerging on-demand service platforms (OSPs) have recently embraced teamwork as a strategy for stimulating workers’ productivity and mediating temporal supply and demand imbalances. This research investigates the team contest scheme design problem considering work schedules. Introducing teams on OSPs creates a hierarchical single-leader multi-follower game. The leader (platform) establishes rewards and intrateam revenue-sharing rules for distributing workers’ payoffs. Each follower (team) competes with others by coordinating the schedules of its team members to maximize the total expected utility. The concurrence of interteam competition and intrateam coordination causes dual effects, which are captured by an equilibrium analysis of the followers’ game. To align the platform’s interest with workers’ heterogeneous working-time preferences, we propose a profit-maximizing contest scheme consisting of a winner’s reward and time-varying payments. A novel algorithm that combines Bayesian optimization, duality, and a penalty method solves the optimal scheme in the nonconvex equilibrium-constrained problem. Our results indicate that teamwork is a useful strategy with limitations. Under the proposed scheme, team contest always benefits workers. Intrateam coordination helps teams strategically mitigate the negative externalities caused by overcompetition among workers. For the platform, the optimal scheme can direct teams’ schedules toward more profitable market equilibria when workers have inaccurate perceptions of the market. History: This paper has been accepted for the Service Science Special Issue on Innovation in Transportation-Enabled Urban Services. Funding: This work was supported by the National Science Foundation [Grant FW-HTF-P 2222806]. Supplemental Material: The online appendices are available at https://doi.org/10.1287/serv.2023.0320 .

Vehicle routing for customized on-demand bus services

This study investigates a variant of the Vehicle Routing Problem (VRP) for customized on-demand bus service platforms. In this problem, the platform plans customized bus routes upon receiving a batch of orders released by passengers and informs the passengers of the planned pick-up and drop-off locations. The related decision process takes into account some passenger-side time window-related requirements, walking limits, the availability and capacities of various types of buses. A mixed-integer linear programming model of this new VRP variant with floating targets (passengers) is formulated. To solve the model efficiently, a solution method is developed that combines the branch-and-bound and column generation algorithms and also includes embedded acceleration techniques such as the multi-labeling algorithm. Experiments based on real data from Dalian, China are conducted to validate the effectiveness of the proposed model and efficiency of the algorithm; the small-scale experimental results demonstrate our algorithm can obtain optimal results in the majority of instances. Additionally, sensitivity analysis is conducted, and model extensions are investigated, to provide customized bus service platform operators with potentially useful managerial insights; for example, a platform need not establish as many candidate stops as possible, a wide range of walking distance may not bring early arrival at destinations for customers, more mini-buses should be deployed than large buses in our real-world case. Moreover, the rolling horizon-based context and zoning strategies are also investigated by extending our proposed methodology.

Efficiency of the carpooling service: Customer waiting and driver utilization

The carpooling service, which is the sharing of journeys heading in similar directions, is expected to be more efficient. In this article, we study the impact of providing a carpooling service on customer waiting time and driver utilization in an on-demand service platform. We build an queueing model to approximate the dynamics of the on-demand service platform that provides two services: a standard service and a carpooling service. We find that two factors influence the efficiency of the carpooling service: the source of customers who use the carpooling service, and the length of a normalized detour. If the carpooling service attracts customers who have not used this on-demand service before, i.e., the market size increases, then both the customer waiting time and driver utilization increase when more customers use the carpooling service. If the carpooling service primarily attracts customers who already use the standard on-demand service before, i.e., customers switch from the standard service to the carpooling service, then the impact of the carpooling service depends on the normalized detour. It is possible that customer waiting time and driver utilization first increase and then decrease as customers switch from the standard service to the carpooling service.

Balancing Agent Retention and Waiting Time in Service Platforms

The gig economy open a new business model in which services have access to a large pool of workers that are compensated based on their actual production, which can be useful to operate at lower levels of utilization to improve response times to customers. However, having a large pool of workers with low utilization may lower their motivation and increase employee turnover, which can hurt productivity in the long run. “Balancing Agent Retention and Waiting Time in Service Platforms” looks at this tradeoff and provides a data-driven approach to manage worker capacity in on-demand service platforms, showing evidence through a real-world application of an outbound call center with freelance agents.

Fast Forecasting of Unstable Data Streams for On-Demand Service Platforms

Fast Forecasting of Unstable Data Streams for On-Demand Service Platforms

Impact of customer bounded rationality on on-demand service platforms

We consider an on-demand platform connecting independent agents and delay-sensitive boundedly rational customers. By adopting a queueing-game framework, we analyze the equilibrium joining behaviors of customers and agents, the platform’s optimal pricing strategy, and the resulting profit and social welfare. Moreover, we investigate how the bounded rationality level affects these results. We obtain the following main insights. First, when customers are irrational, the positive externality between the agents and the customers becomes weak. In other words, increasing the same number of agents attracts fewer boundedly rational customers than fully rational ones. Second, we distinguish two types of equilibria in terms of agent participation: one in which all or none of the agents participate and the other in which it is possible that a fraction of the agents may participate. We show that in the latter case, a social planner and a platform can easily align their goals since they are affected by the bounded rationality in the same direction. We attribute the insight to the interactions among the cross-side effect, bounded rationality, and the pooling effect embedded in a queueing system. Next, we characterize the platform’s optimal pricing decisions and prove that the increase in the bounded rationality level does not always lead to higher optimal prices and wages. Finally, we examine the agent side bounded rationality and the social welfare maximizing decisions in the extended models.

Identification of factors that influence the delivery fee pricing of on-demand delivery services

The use of on-demand delivery services increased in Brazil during the COVID-19 pandemic, mainly by requests for ready meals. While consumers appreciate convenience, the delivery fee is a decisive factor in the purchase process. However, the delivery fee pricing strategy of on-demand delivery service platforms has not been discussed in the literature. Thus, this study aims to analyze the factors that influence the delivery fee pricing of on-demand delivery services and explores the impact of delivery fee strategies on the remuneration of couriers. We collected data from three leading on-demand delivery service platforms regarding product type, order price, service fee, delivery fee, order day, order time, waiting time, and distance. As a result, our database comprises 1,440 orders in 12 Brazilian municipalities. A linear regression model was estimated to identify the factors influencing the delivery fee pricing considering different product types. Findings showed that product type has a diverse effect on delivery fee pricing. Moreover, distance, regardless of the product type, positively influences the delivery fee. The delivery fee of the ready meals is affected by the service fee, waiting time, order day, and order time. Furthermore, the waiting time and order day affect the delivery fee of supermarket and bakery products and beverages. Finally, the delivery fee of medical products is influenced by order day and time. Findings can be helpful for the pricing strategy of on-demand delivery services.

Can two competing on-demand service platforms be profitable?

Can two competing on-demand service platforms be profitable in equilibrium? This question is well-studied for firms competing purely on price in a single-sided market, but it is not well-understood for competing service platforms that use price to influence passenger demand and wage to influence driver participation in a “two-sided” market. To fill this research gap, we analyze the equilibrium structure of different variants of a 2-stage non-cooperative game in which both platforms use lower prices and waiting time to compete for more customers and higher wages and utilization to entice more providers to participate.To explicate our analysis and isolate the effect of each element, we first examine a parsimonious and stylized model that is based on the case when both firms operate under seven operational assumptions: (1) Non-exclusive providers; (2) Non-exclusive customers; (3) Pure pricing strategies; (4) Constant pricing; (5) Homogeneous services; (6) Homogeneous providers; and (7) Homogeneous customers. We find that only one platform can sustain in a payoff dominant stable equilibrium. Then we examine whether this “winner-take-all” equilibrium would persist when those operational environmental assumptions are relaxed separately. Our analysis reveals that the “winner-take-all” phenomenon continues to persist under promotional pricing strategies, service differentiation, and heterogeneous service providers. Interestingly, we find both platforms can sustain profitably when each platform engages providers (or customers) exclusively, when platforms adopt time-based pricing strategies, or when customers are heterogeneous. Our results offer insights into different operating environments under which both platforms can both sustain profitably in equilibrium.

How Crowdworkers Engage in the Age of Algorithms? An Empirical Study of On-Demand Service Platforms

The service industry is experiencing fundamental transformation caused by technological revolutions, namely the emergence of the digital platform and the increasing replacement of managerial functions by algorithms. In this novel service ecosystem – characterised by on-demand fulfilment, algorithmic governance and network effect – crowdworkers play an essential role in providing contingent services for final customers. Since crowdworkers’ work-related positive affect is crucial for sustainable service delivery in platform contexts, this study aims to identify the antecedents of their engagement behaviour. We grounded our theoretical framework in conservation of resources theory and the job demands-resources heuristic model to investigate how the interplay of personal and job resources, as well as job-related demands, could predict crowdworkers’ engagement. We then empirically tested our proposed model among 565 crowdworker respondents. Our results shed light on the engagement behaviour of crowdworkers and highlight the moderator effect of algorithmic justice. Empirical findings also offer pragmatic guidelines for acquiring and retaining a large base of energetic, diligent and engaged service providers that is essential for success and survival in the highly competitive platform market.

How to charge doctors and price medicines in a two-sided online healthcare platform with network externalities?

Platform business models are upending the value proposition of the companies in traditional industries, such as healthcare, through competitively pricing value-added services and products. In this study, we consider a unique, but important, on-demand two-sided service platform: a monopolistic online healthcare platform, which not only provides medical services to patients by connecting them with doctors, but also sells medical products to patients. A crucial operational decision for the platform is to set up a pricing scheme to encourage enough doctors and patients to participate on the platform. We consider two common pricing schemes in this study: fixed and linear pricing, in an empirically-grounded optimisation model that considers medicine pricing decisions together with network externalities. Our results show that the fixed pricing scheme almost always dominates the linear pricing scheme by providing the platform with higher expected profit. Our results also reveal the strong interdependence between the optimal doctor service pricing decision and optimal medicine pricing decision. However, these decisions are often made independently by different functional units of a firm, which could cause suboptimal platform financial performance. These results indicate the importance of pricing products and services in an integrated manner for maximising the profit of a product/service platform.