Online food delivery (OFD) services, known for offering varied meals at home, have gained global popularity. Meituan has recently ventured into the affordable market segment with its "Pinhaofan'' service, highlighting the imperative to delivery efficiency. To achieve this, delivery scope is regarded as one of the most effective operational tools. The delivery scope of a merchant refers to the geo-graphical area where they can serve customers. Current methods for generating delivery scopes primarily focus on optimizing a single merchant's efficiency or rely on manual delineated from the merchant's perspective, neglecting the merchant substitution effect and potentially resulting in order loss. In this paper, we propose a novel method, named Collaborative Scope, which views the delivery scope as an assortment optimization problem, considering the substitution effect between merchants from the user's perspective. We introduce the discrete choice model of econometrics and use the Enhanced Multinomial Logit Model to predict user conversion rates in the merchant list. Next, we formulate the delivery scope optimization problem of multiple merchants as a mixed integer programming problem. The city-wide solution of this problem, owing to the large-scale combinatorial optimization triggered by high-dimensional decision variables, incurs high computational complexity. To address this, we propose an approximate solution to the original problem through a first-order Taylor series approximation, which significantly reduces the computation complexity at the expense of a slight decrease in solution accuracy. Offline and online A/B test results indicate that, compared to existing methods, Collaborative Scope significantly improves delivery efficiency by reducing delivery difficulty without hurt of order volume. Notably, Collaborative Scope is currently deployed on "Pinhaofan'', serving tens of millions of online users.

Online food delivery (OFD) services have become popular globally, serving people's daily needs. Precise area-of-interest (AOI) boundaries help OFD platforms determine customers' exact locations, which is crucial for maintaining consistency in delivery difficulty and providing a uniform customer experience within an AOI. Existing AOI generation methods primarily rely on predefined shapes or density-based clustering, which limits the quality of the contours. Recently, Meituan has treated the AOI contours as a binary semantic segmentation problem. Their approach involves a multi-step post-process to address the issues with boundary breaks caused by semantic segmentation models, leading to decreased quality and inefficiency in the learning process. In this paper, we propose a novel method for AOI contour generation called C-AOI (Contour-based Area-of-Interest). C-AOI is an instance segmentation model that focuses on generating high-quality AOI contours. Unlike the former method, which relies on pixel-by-pixel classification, C-AOI starts from the center point of the AOI and regresses the boundary. This approach results in a higher-quality boundary and is less computationally intensive. C-AOI first corrects errors on the contour using a local aggregation mechanism. Then, we propose a novel deforming module called the contour transformer, which captures the global geometry of the object. To enhance the positional relationship among vertices, we introduce a learnable cyclic positional encoding applied to the contour transformer. Finally, to improve the boundary details, we propose the Adaptive Matching Loss (AML) that eliminates over-smoothed boundaries and promotes optimized convergence pathways. Experimental results on real-world datasets collected from Meituan have demonstrated that C-AOI significantly improves the mask and boundary quality compared to Meituan's previous work. Moreover, Its inference speed is comparable to that of E2EC, a state-of-the-art real-time contour-based method. It is noteworthy that C-AOI has been deployed in the Meituan platform for producing AOIs.

On-demand food delivery (ODFD) is booming globally. ODFD services rely heavily on accurate Areas of Interest (AOIs) data to make effective operational decisions, such as service areas of restaurants. Few methods can automatically generate AOIs with contours highly matched to geographic boundaries, so manual labeling and auditing still play a vital role in data production. Most existing studies can merely generate approximate ranges which are not accurate enough for business usage. The others, which can generate AOIs fitting geographic boundaries through map matching methods, are limited by the deficiency of road network data, as geographic boundaries contain rivers, mountains, and so on. To address this issue, we propose a novel AOI generation framework using large-scale geospatial data to generate AOIs which are closer to geographic boundaries. In our framework, we firstly extract multimodal features from satellite images, road networks as well as delivery data (customer addresses, locations, etc.), and then use a semantic segmentation model to infer pixel-level points that possibly lie within the boundaries of AOIs. After that, a contour learning method and simple post-processing are applied to fit the discrete pixel-level points to reconstruct contours of AOIs in arbitrary shapes. Experiments were conducted aiming at comparing the proposed framework with six competing methods qualitatively and quantitatively. As a result, our proposed framework performs better in generating AOIs with accurate and geometry-preserving contours.

With the development of mobile Internet, online food delivery (OFD) services have become increasingly popular in our daily lives. OFD platforms rely heavily on accurate Areas-of-Interest (AOIs) information on many aspects of their operations to pinpoint customers' exact locations and to define the service areas of restaurants. Recently, OFD platforms have started to tap into the vast amount of geospatial data generated in their day-to-day business to improve the accuracy of their AOI information. Although there has been a proliferation of studies that leverage such data to detect the underlying AOIs, for example, to identify the names and spatial boundaries of the AOIs, they focus on the single-AOI detection problem, that is, they detect AOIs one at a time and ignore their spatial dependency. This would end up with inconsistent results, i.e., AOIs with overlapping spatial boundaries. To address this issue, we propose a new approach to detect multiple AOIs simultaneously and solve the multi-AOIs detection problem. In our approach, we first apply the existing single-AOI detection algorithms to generate candidate spatial boundaries for AOIs in a neighborhood, and then develop a Binary Integer Linear Programming (BILP) model to determine the best candidate spatial boundaries for these AOIs while accounting for their spatial dependency. We conduct numerical experiments using real data from Meituan, the largest OFD platform in China. Results show that our model not only produces consistent AOI boundaries, but also improves the average F1 score by 4.7%.

Large amounts of efforts have been devoted into learning counterfactual treatment outcome under various settings, including binary/continuous/multiple treatments. Most of these literature aims to minimize the estimation error of counterfactual outcome for the whole treatment space. However, in most scenarios when the counterfactual prediction model is utilized to assist decision-making, people are only concerned with the small fraction of treatments that can potentially induce superior outcome (i.e. outcome-oriented treatments). This gap of objective is even more severe when the number of possible treatments is large, for example under the continuous treatment setting. To overcome it, we establish a new objective of optimizing counterfactual prediction on outcome-oriented treatments, propose a novel Outcome-Oriented Sample Re-weighting (OOSR) method to make the predictive model concentrate more on outcome-oriented treatments, and theoretically analyze that our method can improve treatment selection towards the optimal one. Extensive experimental results on both synthetic datasets and semi-synthetic datasets demonstrate the effectiveness of our method.

A Point-Of-Interest (POI) represents a specific point location that may be useful or interesting for people, and therefore each and every building footprint in a topographic map can be recognized as a POI. Automatic extraction of building footprints using remote sensing images has become a challenging and important research topic which is in demand for urban planning and development. Extensive studies have explored a variety of semantic segmentation methods using deep learning algorithms to achieve better performance in building footprint extraction, however the existing algorithms were shown to have some limitations which lead to poor segmentation results. Building roofs were recognized as building footprints in the previous studies. This is prone to error especially for high-rise buildings due to different sensor view angles. In this paper, we propose a multi-task Res-U-Net model with attention mechanism for the extraction of the building roofs and the whole building shapes from remote sensing images, then use an offset vector method to detect the footprints of the high-rise buildings based on the boundaries of the corresponding building roofs and shapes. We also apply the online food delivery (OFD) data to parse the POI name of every building footprint. Several strategies are also developed in combination with the proposed model, including data augmentation and post-processing. We conduct numerical experiments using real data of remote sensing images and OFD historical order data. Results demonstrate that our proposed model achieves a total F1-score of 77.05% and intersection over union (IoU) of 63.55% in terms of the building roof segmentation, and an overall F1-score of 79.02% and IoU of 66.05% for the whole building shape segmentation, which both achieve the best performance among all baseline models.

Nowadays, on-demand food delivery service has become fashionable in China. The efficiency of food delivery relies heavily on accurate coordinates of destination Points of Interest (POI). However, the coordinates of the destination POIs from the existing geospatial data warehouses still have many problems that perplex couriers severely. The major problems can be concluded in two categories: 1) the deviation of POI coordinates; 2) the lack of POI coordinates. To address these problems, we propose a POI-coordinate-generating framework based on couriers' and users' behavioral data of historical waybills. In particular, we start with a combinatorial strategy to assign waybills to Areas of Interest (AOI). Second, we generate a destination POI name by processing the user address for each waybill, and all waybills are grouped by the corresponding POI name. Then, a data density image of the behavioral data is generated for each group, with the ground-truth location of the POI labeled. Finally, a U-Net is trained by using the images generated in the previous step to infer locations of the POIs. We evaluated this framework by launching experiments and case studies on large-scale datasets, and the result shows our framework can predict coordinates of POIs accurately. These predicted coordinates can be used to calibrate deviated coordinates of many POIs and complement the geospatial data warehouse.