Stefanus Jasin

Stefanus Jasin

Auto-generated profile, based on publicly available information

Stefanus Jasin

Ross School of Business, University of Michigan, Ann Arbor, Michigan 48109

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 September 2019 in Information Systems Research
Verified
Journal Article
Closed access
YHYan Huang
Yan Huang
SJStefanus Jasin
Stefanus Jasin
PMPuneet Manchanda
Puneet Manchanda
We propose a novel two-stage data-analytic modeling approach to gamer matching for multiplayer video games. In the first stage, we build a hidden Markov model to capture how gamers' latent engagement state evolves as a function of their game-play experience and outcome and the relationship between their engagement state and game-play behavior. We estimate the model using a data set containing detailed information on 1,309 randomly sampled gamers' playing histories over 29 months. We find that high-, medium-, and low-engagement-state gamers respond differently to motivations, such as feelings of achievement and need for challenge. For example, a higher per-period total score (achievement) increases the engagement of gamers in a low or high engagement state but not those in a medium engagement state; gamers in a low or medium engagement state enjoy within-period score variation (challenge), but those in a high engagement state do not. In the second stage, we develop a matching algorithm that learns (predicts) the gamer's current engagement state on the fly and exploits that learning to match the gamer to a round to maximize game-play. Our algorithm increases gamer game-play volume and frequency by 4%–8% conservatively, leading to economically significant revenue gains for the company.

1 April 2023 in Management Science
Verified
Journal Article
Closed access
QWQiaochu Wang
Qiaochu Wang
YHYan Huang
Yan Huang
SJStefanus Jasin
Stefanus Jasin
PSParam Vir Singh
Param Vir Singh
Should firms that apply machine learning algorithms in their decision making make their algorithms transparent to the users they affect? Despite the growing calls for algorithmic transparency, most firms keep their algorithms opaque, citing potential gaming by users that may negatively affect the algorithm’s predictive power. In this paper, we develop an analytical model to compare firm and user surplus with and without algorithmic transparency in the presence of strategic users and present novel insights. We identify a broad set of conditions under which making the algorithm transparent actually benefits the firm. We show that, in some cases, even the predictive power of the algorithm can increase if the firm makes the algorithm transparent. By contrast, users may not always be better off under algorithmic transparency. These results hold even when the predictive power of the opaque algorithm comes largely from correlational features and the cost for users to improve them is minimal. We show that these insights are robust under several extensions of the main model. Overall, our results show that firms should not always view manipulation by users as bad. Rather, they should use algorithmic transparency as a lever to motivate users to invest in more desirable features. This paper was accepted by D. J. Wu, information systems. Supplemental Material: The online appendix is available at https://doi.org/10.1287/mnsc.2022.4475 .

1 September 2024 in Operations Research
Verified
Journal Article
Closed access
QCQi (George) Chen
Qi (George) Chen
YLYanzhe (Murray) Lei
Yanzhe (Murray) Lei
SJStefanus Jasin
Stefanus Jasin
In “Real-Time Spatial–Intertemporal Pricing and Relocation in a Ride-Hailing Network: Near-Optimal Policies and The Value of Dynamic Pricing,” Chen, Lei, and Jasin consider a dynamic pricing problem faced by a ride-hailing service provider who manages a fixed number of servers and serves price-sensitive customers within a network. Servers serve arriving customers by relocating from the requested origins to destinations within a certain travel time. The authors first propose a static pricing policy based on the optimal solution to a deterministic relaxation of the original stochastic problem. They show that the proposed static policy matches the best possible asymptotic performance of any static policy. The authors further propose a dynamic pricing policy that adaptively changes the prices in a way that reduces the impact of past demand randomness on the balance of future distributions of servers and customers across the network. They show that the dynamic pricing policy achieves significantly better asymptotical performance. The proposed policies and their performance guarantees are further extended to a case where the firm jointly decides the relocation of vacant servers

1 November 2021 in Manufacturing & Service Operations Management
Verified
Journal Article
Closed access
LWLai Wei
Lai Wei
RKRoman Kapuscinski
Roman Kapuscinski
SJStefanus Jasin
Stefanus Jasin
Problem definition: Shipment consolidation (i.e., shipping multiple orders together instead of shipping them separately) is commonly used to decrease total shipping costs. However, when the delivery of some orders is delayed, so they can be consolidated with future orders, a more expensive expedited shipment may be needed to meet shorter deadlines. In this paper, we study the optimal consolidation policy focusing on the trade-off between economies of scale due to combining orders and expedited shipping costs, in the setting of two warehouses. Academic/practical relevance: Our work is motivated by the application of fulfillment consolidation in e-commerce and omni-channel retail, especially with the rise of so-called on-demand logistics services. Sellers have the flexibility to take advantage of consolidation by deciding when to ship the orders and from which warehouse to fulfill the orders, as long as the orders’ deadlines are met. Methodology: We use Dynamic Programming to study the optimal policy and its structure. We also conduct extensive simulation tests to evaluate the performance of heuristics that are based on structures of the optimal policies. Results: The optimal policies and their structures are characterized. Using the insights of these structural properties, we propose two easily implementable heuristics that perform within 1%–2% of the optimal solution and outperform other benchmark consolidation methods in numerical tests. Managerial implications: Consolidation is shown to significantly reduce the outbound shipping costs. Retailers can take advantage of it to effectively improve the standard policies by simply applying the threshold-form heuristics we propose.

1 January 2022 in Operations Research
Verified
Journal Article
Closed access
SMSentao Miao
Sentao Miao
SJStefanus Jasin
Stefanus Jasin
XCXiuli Chao
Xiuli Chao
Simple Algorithms for Complex Multiwarehouse, Multistore Inventory Control Problems Retailers (both brick-and-mortar and e-commerce) have always faced the problem of allocating inventories in their warehouses (or central distribution centers) to the stores (or smaller local warehouses) in order to minimize total costs. The problem is particularly challenging when the network structure is large and complex, the selling season is long, and the replenishment is frequent. For example, giant retail chains such as Macy’s typically have many warehouses and hundreds of stores across the United States, and online retailers such as Amazon have many distribution centers and over one hundred fulfillment centers. The authors develop algorithms to solve this multiwarehouse, multistore (MWMS) inventory control problem. Their algorithms are computationally efficient and asymptotically optimal as the problem becomes large and complex. This feature is very appealing to today’s fast-moving retail industry with rapidly expanding business scale.

1 October 2022 in Management Science
Verified
Journal Article
Closed access
XGXiangyu Gao
Xiangyu Gao
SJStefanus Jasin
Stefanus Jasin
SNSajjad Najafi
Sajjad Najafi
HZHuanan Zhang
Huanan Zhang
We consider joint learning and optimization problems under a general Cascade Click model. Under this model, customers examine the products in a decreasing order of display, from the top to (potentially) the bottom of the list. At each step, customers can decide to either purchase the current product, forego the current product and continue examining the next product, or simply terminate the search without purchasing any product. We first consider the core pricing problem, where the display position (ranking) of each product is fixed and the only decision that the firm needs to make is pricing. We then consider an extension to the problem of joint ranking and pricing in the presence of filtering options, which the customers can use to filter out some undesirable products. For both problems, we develop Upper Confidence Bound (UCB)-based joint learning and optimization algorithms with theoretical performance guarantees. The key challenge here is in constructing a UCB algorithm that exploits the structure of the Cascade Click model while at the same time taking into account all of the historical click and purchase information. Our numerical results yield three key insights. First, naively applying a standard black box UCB algorithm without adapting it to the Cascade structure is very inefficient and results in a huge loss in total revenues during a finite horizon. Second, applying a learning algorithm by assuming a mis-specified model that ignores the Cascade behavior may result in a highly suboptimal solution. Third, jointly optimizing ranking and pricing can significantly improve performance. Thus, although in practice these decisions are sometimes made separately due to organizational structure, our results suggest that a significant benefit can be realized when the two decisions are more closely coordinated. This paper was accepted by J. George Shanthikumar, big data analytics.