Operations Research grad student helps Newark optimize trash collection

January 3, 2011 under CANR News

Priyanka Jain, a master’s degree student in the College of Agriculture and Natural Resources, is working with the city of Newark to study ways to optimize residential trash pick-up and save costs.

Jain, who is in CANR’s operations research program, explained that the main goal of her work is to “enhance waste collection practices in the city of Newark in terms of minimizing fleet size, total transportation and operational cost, and avoiding time imbalance in between different routes.”

The study has two main parts. First, Jain looked at assigning different capacity trucks to various routes to help cut down on the number of trips taken by each truck. The city has trucks of varying capacity and Jain saw that specific types of trucks worked better on certain routes.

Jain found that a smaller model of truck was making two trips to pick up the same amount of waste that could be handled by a larger truck in one trip. She said she would like to cut the number of trips to save on fuel, operational costs and overtime pay.

Because there is less trash to pick up in the winter, Jain said she believes the city can collect all the trash successfully with four trucks rather than the five they currently use.

By decreasing the number of trips taken by each truck on their routes, Jain’s research showed a 19 percent reduction in yearly transportation and drivers’ labor costs.

The second part of the study concerned route optimization to save on fuel and overtime costs.

To determine the optimal route depending on the average waste to be collected, Jain used Network Analyst, an ArcGIS extension for problems such as shortest route, closest facility, location allocation and vehicle routing.

Jain said of the city’s current routing plan, “They have a good scheme, but still there are some trucks that have to do multiple trips because there are uncovered remaining houses. I’m trying to make routes, different routes, so that they have very optimal collection schemes and they don’t have to go back.”

Using optimal route solutions for the city, the ArcGIS computed using traffic directions, turn restrictions, average speeds for local roads and highways and average time for serving each bin. It included geocoding of the city’s customers on GIS maps, which can be helpful in the future if more customers need to be added. City historical data was used to calculate average drop off time at the transfer station, the area where the trucks transfer their waste. Field observations were also conducted to assess the average turn times and service time for bins.

When these optimized routes were compared to the current ones, the results showed that distance would be decreased between 4-15 percent on each route, with an average of a 9 percent reduction in mileage, leading to an estimated decrease of fuel costs by $1,500 and maintenance costs of $7,000 per year per route.

Cost is not the only benefit from Jain’s research, however, as she says another plus that comes from route optimization will be public safety.

Jain said she is “trying to optimize their routes so they do fewer U-turns, which is critical in terms of safety. They are huge trucks and when they back up, if they make a three-point turn, it is a main concern especially in terms of safety. They don’t want the trucks to make many U-turns or three-point turns.”

With fewer trucks running more efficient routes, there will be an environmental benefit to the research as well, as fewer trucks driving fewer miles will help Newark reduce its carbon footprint.

The study originated in a class taught by Kent Messer, assistant professor of food and resource economics and assistant professor of economics, and Messer says Jain was “just a wonderful example of someone going above and beyond and demonstrating her passion and knowledge. She obviously did a great job.”

Messer also said that the city of Newark was very helpful to Jain throughout her research. “They are a great team, and I give them kudos for doing it because they have to get a lot of data to run these things,” he said. “They’re very data intensive to get good meaningful results. So I just think that it’s a beautiful relationship between the University of Delaware and a student and the city.

“I think her analysis was great, and the thing that I like about it is that I think they’re going to do it. From what I can tell, they’re going to go try it out, run some of these routes, get feedback and see whether it’s actually going to get put on the ground. And that’s so much better than a study by itself.”

Along with Messer, Jain credited Rich Lapointe, the director of public works for the city; Patrick Bartling, public works superintendent for providing a lot of support, information and data; and Benjamin Mearns, information resources consultant with the University’s IT-Client Support and Services, for helping her with ArcGIS.

Jain will continue her study into next semester, adding things such as more detailed traffic data and recycling into her analysis.

Article by Adam Thomas
Photo by Danielle Quigley

This article can also be viewed online on UDaily by clicking here.

Share

CANR Students Conduct Community Service

November 8, 2010 under CANR News

CANR students have been spending time this fall volunteering for community service organizations.

In October, Alpha Zeta went to the Brandywine Zoo for two hours for our community service project. Most of the members raked leaves to help keep the zoo looking neat. The leaves that we raked were also used for bedding for the animals in the winter. Other members helped renovate an animal holding pen area.

UD students in Kent Messer’s FREC100 class volunteered for The Nature Conservancy on November 7th.  “This is the second year where my students have volunteered with The Nature Conservancy and it appears to be developing into a long-term partnership as its seems to be beneficial to all involved,” Messer said.

Share

Conservationists model smart shopping, save big

February 3, 2010 under CANR News

If there was ever a time for conservationists to shop smart, this is it. Across America, states confront budget shortfalls — a grim cycle of cuts, followed by more cuts that the Pew Center on States refers to as “fiscal crisis.” Tough choices confront land conservationists, who shop the American landscape with big ambitions but slim wallets.

That’s why an emerging economics tool is so timely. Researchers at the University of Delaware and The Conservation Fund have designed a computer based decision making tool that is helping conservationists get more bang for their buck — by evaluating potential conservation projects for best dollar value. With this new tool, government leaders can comparison-shop projects like never before.

“We all want the most bang for our buck, and conservation is no different,” says Will Allen, director of strategic conservation at The Conservation Fund, a leading environmental nonprofit. “Are you spending too much money on expensive projects — what some call budget sponges — or are you getting real value? With public budgets so tight, government officials must be able to justify how they’re spending these dollars wisely.”

Until now, conservation organizations have chosen which lands to conserve based primarily on land benefits. Their goal is to save high-priority land that’s valued for agriculture, perhaps, or open space and wildlife habitat. But Baltimore County, Md., has a better way, using optimization techniques developed by UD’s Kent Messer, an economist in the College of Agriculture and Natural Resources, and Allen.

According to Messer, optimization strategy, first used by the military during World War II, is gaining new ground in the conservation world. He says it would have once taken a supercomputer to run these calculations. But thanks to evolving technology, people can now run optimization analysis quickly for fields like conservation, which has largely operated more on a wish-list level.

Messer compares buying land to buying wine — choosing well means recognizing good value when you see it.

Messer and Allen have developed and applied a computer model to “optimize” conservation decisions. The model turns raw data about conservation decisions — project costs, benefits (scored numerically), budget constraints — into a user-friendly spreadsheet yielding comparison shopping conclusions.

Using the model, for example, a government agency can quickly compare the relative value of all possible projects — and then make, and justify, an informed choice.

The Baltimore County Agricultural Land Preservation Program in Maryland has one of the nation’s most well-established farmland preservation efforts. Every year since 2007, Baltimore County has tapped an optimization model to choose which agricultural lands to save with impressive results.

Over the past three years, Baltimore County staff estimate that optimization has helped the county protect an additional 680 acres of high-quality agricultural land at a cost savings of roughly $5.4 million compared to the class conservation decisions tools. This amounts to a return on investment of more than 60 to 1. In other words, for every dollar that Baltimore County spent using its optimization model, it has gained more than $60 in conservation benefits.

“This work is especially important in these times of constrained budgets. Making our money go as far as possible is a good thing,” says Michael McGrath, chief of planning for the Delaware Department of Agriculture (DDA). Messer used data provided by the DDA for some of his initial research related to optimization. He and McGrath continue to look for ways that UD and DDA can use these techniques in Delaware.

McGrath says, “Dr. Messer’s current work in Baltimore County provides important substantiation of his techniques in optimization. These techniques have applicability in Delaware and all across the U.S. in optimizing farmland preservation easement negotiations.”

Read the article here on UDaily.

Share