A New Formulation for the Dial-a-Ride Problem

This paper proposes a new mixed integer programming formulation and branch and cut (BC) algorithm to solve the dial-a-ride problem (DARP). The DARP is a route-planning problem where several vehicles must serve a set of customers, each of which has a pickup and delivery location, and includes time window and ride time constraints. We develop “restricted fragments,” which are select segments of routes that can represent any DARP route. We show how to enumerate these restricted fragments and prove results on domination between them. The formulation we propose is solved with a BC algorithm, which includes new valid inequalities specific to our restricted fragment formulation. The algorithm is benchmarked on existing and new instances, solving nine existing instances to optimality for the first time. In comparison with current state-of-the-art methods, run times are reduced between one and two orders of magnitude on large instances.

Does Delivery Location Matter? A National Study of the Impact of Dual Enrollment on College Readiness and Early Academic Momentum

Background/Context Although dual enrollment (DE) programs have indicated positive impact on various high school and postsecondary outcomes, access to DE programs remains unequal; historically marginalized students are less likely than other students to attempt college credits in high school. Despite DE being a widely adopted program at the state level, these programs vary greatly by eligibility criteria, funding models, delivery location, and modality. Purpose/Objective/Research Question/Focus of Study Guided by prominent learning theories, we hypothesize that the influence of early DE on later educational pathways and outcomes may vary by the location in which DE is delivered. This study examines whether the delivery location of DE (i.e., on a college campus or otherwise) influences students’ college readiness and first-year academic momentum in college, with a special focus on its heterogeneous effect among students of diverse racial and socioeconomic background. Research Design Using the restricted-use data from High School Longitudinal Study of 2009 (HSLS:09), we use a quasi-experimental approach (i.e., inverse probability weighting models) with a nationally representative sample of students who have taken at least one DE course by 11th grade. Findings/Results The findings reveal that students who took at least one DE course on a college campus do not differ in their cumulative high school GPA, in their probability of attending college, in whether they took developmental courses, in whether they attended college immediately after high school graduation, and in their probability of full-time enrollment when compared with those who took DE course(s) elsewhere. However, the findings are not applicable to all students of varying background defined by race/ethnicity and socioeconomic status. Conclusions/Recommendations This study provides several implications: (1) Because DE courses taken on a high school or college campus equally fuel students’ college readiness and early academic momentum, advising practices should acknowledge the benefits of DE courses regardless of delivery location. (2) DE participation with college exposure may particularly benefit students of higher socioeconomic status (SES), so interventions that offer holistic college experiences beyond academic work are needed to effectively prepare lower SES students for college life and accumulate academic momentum are needed. (3) States and educational entities should be mindful about the potential disparate effect of DE programs and provide regulation, oversight, and quality assurance so that these programs can narrow the postsecondary achievement gap.

Order Fulfillment Decision under Multiwarehouse Collaborative Delivery

Aiming at the high cost of multicategory orders fulfillment under multiwarehouse collaborative distribution, comprehensively considering the fulfillment costs of different orders fulfillment strategies, an order fulfillment strategy selection model is proposed. The first step of the model uses the linear programming algorithm to solve the cost of suborder merge transportation after the order is split. The second step calculates the cost of the current “greedy algorithm” of the e-commerce platform for order split fulfillment. Then, the cost of each strategy is compared and the lowest cost one is chosen. The calculation example analysis shows that the order fulfillment strategy is closely related to the delivery location of the order and the SKU category. When the delivery location is far away and SKU categories are many in the order, the merged transportation strategy of suborders after the order is split will be significantly better than the cost of separate transportation. The multiwarehouse collaborative distribution fulfillment strategy proposed in this paper can provide a decision basis for the e-commerce platform to choose which fulfillment method.

Closing the gap on institutional delivery in northern India: a case study of how integrated machine learning approaches can enable precision public health

IntroductionMeeting ambitious global health goals with limited resources requires a precision public health (PxPH) approach. Here we describe how integrating data collection optimisation, traditional analytics and causal artificial intelligence/machine learning (ML) can be used in a use case for increasing hospital deliveries of newborns in Uttar Pradesh, India.MethodsUsing a systematic behavioural framework we designed a large-scale survey on perceptual, interpersonal and structural drivers of women’s behaviour around childbirth (n=5613). Multivariate logistic regression identified factors associated with institutional delivery (ID). Causal ML determined the cause-and-effect ordering of these factors. Variance decomposition was used to parse sources of variation in delivery location, and a supervised learning algorithm was used to distinguish population subgroups.ResultsAmong the factors found associated with ID, the causal model showed that having a delivery plan (OR=6.1, 95% CI 6.0 to 6.3), believing the hospital is safer than home (OR=5.4, 95% CI 5.1 to 5.6) and awareness of financial incentives were direct causes of ID (OR=3.4, 95% CI 3.3 to 3.5). Distance to the hospital, borrowing delivery money and the primary decision-maker were not causal. Individual-level factors contributed 69% of variance in delivery location. The segmentation analysis showed four distinct subgroups differentiated by ID risk perception, parity and planning.ConclusionThese findings generate a holistic picture of the drivers and barriers to ID in Uttar Pradesh and suggest distinct intervention points for different women. This demonstrates data optimised to identify key behavioural drivers, coupled with traditional and ML analytics, can help design a PxPH approach that maximise the impact of limited resources.