SOLID WASTE COLLECTION

ROUTING OPTIMIZATION:

THE CITY OF VIRGINIA BEACH

STORY

Robert B. Gardner, PE,

BCEE

Senior Vice President

SCS Engineers

Norfolk, Virginia

John C. Barnes, PE

Waste Management

Administrator

Virginia Beach, Virginia

Kevin Callen

Principal

Route Optimization

Consultants

Manassas, Virginia

Overview

• Background

• Demographics and Approach

• Routing Evaluations and Results

Background• Why was route optimization needed?

– Early completion times due to:

• Reduction in tonnage

• Increased collection capacity due to switch

from single axle to tandem

– Workload imbalance

– Stewards of the citizens money

– Potential for privatization

Background

• Prior to 2008 - Route boundaries

understood by operators; but not

drawn

• 2008 - Manually balanced routes

• 2012 –Route Optimization and Right

Size Study initiated

Background

• Routing Options

– Don’t Route or balance

– Manual balance routes

– Route Optimization

• Perform In house

– Limited staff availability and computer expertise

• Contract

– Expertise with collections, software, data analysis,

safety, equipment, and industry practices

– Improved confidence in results

Demographics and Approach

• Demographics

• Collection Service

Rural areas

Beach areas

Urban areas

Demographics and Approach

• Demographics

• Collection Service

• Pre-study statistics

– Equipment mix

– Coordinator boundaries

Tandem Axel– Red

Single Axel - Green

Route Coordinator Areas of

5-8 Routes Each

Original Routes

Day Time (hrs) Stops Containers Mileage

Total

Tons

Dump

Trips

Number

of Trucks

Average

Time per

Route

(hrs)

Tuesday 356.1 30,185 35,530 2,461 587 116 39 9.1

Wednesday 344.5 31,048 34,794 2,266 604 114 38 9.1

Thursday 317.5 31,313 34,689 1,259 609 115 38 8.4

Friday 337.2 31,525 34,790 1,925 613 115 38 8.9

Total 1,355.2 124,071 139,803 7,911 2,414 460 153 35.4

Average 338.8 31,018 34,951 1,978 604 115 38 8.9

Range 38.6 1,340 841 1,202 26 2 1 0.8

0 50 100 150 200 250 300 350 400

Tuesday

Wednesday

Thursday

Friday

Hours

Three Phase Implementation

Plan• Phase I – 2013 - Complete

• 22 tandem trucks and 13 single axle trucks

• 35 routes per day

• 27 total tandem trucks in fleet

• Phase II – 2014 – Complete • 27 tandem trucks and 5 single axle trucks

• 32 routes

• 32 total tandem trucks in fleet

• Phase III – 2015 – Under Design• Estimated 30 routes per day (Tuesday with 5

single axles; Friday with none)

• 37 total tandem trucks in fleet

Data Set-up Between Phases 1,

II, & III

• Generating Baseline Statistics for

Comparison

– Phase I

– Phases II & III

Phase I

• Route for mix of Tandem and Single

Axle

• Model calibrated to match existing

production rates

• Conducted area routing and path

routing

Phase I Results

• Reduced weekly routes from 153 to

143

• 11% reduction in route hours

• 12% reduction in mileage

• 17% reduction in dump trips

• Path routing done, but not used

• Operational adjustments were

“bumpy”

Phase II

• Route for additional Tandems

• Did not do complete re-route

• Improved accuracy of address data

• Improved weight ticket data

• GPS installed and available

Phase II Results

• Reduced routes from 143 to 128

• Additional 9% reduction in the number

of routes

• Additional10% reduction in labor hours

• Improved mapping

• Operators easily adapted

3rd and Final Phase

(Under Design)

• Review Day boundaries for

workload imbalance and future

development

• All required tandems available

• Complete re-route

• Safety and limited access streets

inventoried

Comparison of Route

Optimization Software

• Common Features– Did not create highly usable travel paths

– Provided unique modeling of the waste routes

• FleetRoute– No historical route productivity data needed

– Better handling of mixed density areas with rural and

suburban neighborhoods

• WMDesign– Processed route solutions significantly faster

– More accurate setup using actual productivity rates

Safety Improvements

• Identified limited access streets

• Established routes for dedicated back-

down truck with spotter

• Identified streets with service time

constraints

• Some single axles required to service

limited access and backing streets

Additional Tasks

• Routing for bulk item collections (point

to point)

• Yard debris routes; vary by

subdivision, tree cover, grass, season,

etc

Lessons Learned

• Safety Improvements

• Change is hard; expect resistance

• Anticipate ripple effects as much as

possible

• Need good maps

• Path routing not useful

Overall Results

• Reduced routes from 38 to 30

• Cost savings :

– $2,000,000 reduction in equipment

inventory

– $784,000 annual savings in personnel,

equipment maintenance and fuel

• Completion time

• City leadership and Collection Fee

Customers