Use/Demand
-
Upload
carrington -
Category
Documents
-
view
28 -
download
0
description
Transcript of Use/Demand
Use/Demand
• Concepts/units
• Methods
• Use in Planning– Management/operations– Evaluation– Needs assessment– Forecasts
Why Measure Use
• Operational decisions– maintenance, staffing, budget allocations
• Needs
• Performance measure/evaluation
What to measure
• Use units – individual, groups…visit, visitor, trip, day, RVD, vehicle entry,…
• Level of detail/aggregation– Temporal– Spatial– By activity– By user characteristics
Use estimation questions
• Measure or estimate use at existing site
• Forecast use at existing site
• Forecast changes in use– New sites/products– Change at site or within the environment
Key determinants of use• Population size & characteristics• Distances to markets• Tastes & preferences• Prices• Substitutes• Quality/quantity of supply• Promotion/information• Weather/climate
Methods for estimating use
• On-site techniques– Counting devices– Observation– Surveys (sample or census)
• Off-site– Use prediction Models – e.g. Gravity model
• Trip generation
• Trip distribution
Double Sampling Technique
• 1. Select a variable that is easily measured• E.g. water meter - W
• 2. Establish a relationship with use• U = a * W + c
• Calibration via “double sampling” /linear regression
• U = 1/3 * W - 2
• 3. Regularly measure W and apply equation
• 4. Periodically recalibrate
See page 21 of lecture handouts for details
Sampling to estimate use
• Stratify time periods into H, M, L use• E.g. weekend, weekday, am, pm, eve
• Sample representative periods within each stratum
• Measure use during each sampled period, compute averages
• Expand from sample to population
Lansing Trail Strata: Use counts for one hour
Day Am (8-12) Pm (12-4) Eve (4-8)Weekday 7,8,18
Avg =11
3,8,8,8, 101
Avg= 25
78,59,139,13,72
Avg=72
Weekend 11, 21, 9
Avg= 14
8, 6, 116
Avg =43
8,104,123
Avg =135
Overall avg = 42,Total use = 42 *12 hours*31 days = 15,624
Gravity Model
• Ti,j = K* Pi *Aj / D2ij
• Where
– Ti,j = trips from origin i to destination j
– Pi = population of origin I
– Aj = measure of attractiveness of destination j
– Dij = distance from i to j (or cost)
– K =calibration constant
Linear regression model
• V = a0 + a1 X1 + a2 X2 + a3 X3 +… an Xn
• Where
– a’s are coefficients to be estimated– X’s are independent variables/use predictors– E.g. V = participation rate
» Frequency of participation
» Visits in past year
– X’s : age, income, gender, location, weather…
Demography
• Forecast size, geographic distribution, and age-sex composition of the population
• Census 2000 data
• American Fact Finder
Participation by demographic characteristics
• Age
• Income
• Gender
• Rural/urban, location
• Racial/ethnic group