Ensemble machine learning and forecasting can achieve 99% uptime for rural handpumps.

Broken water pumps continue to impede efforts to deliver clean and economically-viable water to the global poor. The literature has demonstrated that customers' health benefits and willingness to pay for clean water are best realized when clean water infrastructure performs extremely well (>99% uptime). In this paper, we used sensor data from 42 Afridev-brand handpumps observed for 14 months in western Kenya to demonstrate how sensors and supervised ensemble machine learning could be used to increase total fleet uptime from a best-practices baseline of about 70% to >99%.

Measuring and Increasing Adoption Rates of Cookstoves in a Humanitarian Crisis.

Traditional smoky cooking fires are one of today's greatest environmental threats to human life. These fires, used by 40% of the global population, cause 3.9 million annual premature deaths.

Reducing Ultrafine Particle Emissions Using Air Injection in Wood-Burning Cookstoves.

In order to address the health risks and climate impacts associated with pollution from cooking on biomass fires, researchers have focused on designing new cookstoves that improve cooking performance and reduce harmful emissions, specifically particulate matter (PM). One method for improving cooking performance and reducing emissions is using air injection to increase turbulence of unburned gases in the combustion zone. Although air injection reduces total PM mass emissions, the effect on PM size distribution and number concentration has not been thoroughly investigated.

Bending characteristics of polymethylmethacrylate columns, connecting bars of carbon fiber, titanium, and stainless steel used in external skeletal fixation and an acrylic interface.

Objective: To (1) mechanically evaluate polymethylmethacrylate (PMMA) columns of various sizes and compare them to connecting bar materials (carbon fiber composite, titanium, stainless steel) and (2) compare the properties of an intact PMMA column to those of an acrylic interface.

Study Design: Experimental mechanical study.

Sample Population: Experiment 1: 6 groups of 6 specimens each; experiment 2: 2 groups of 12 specimens each.