Date post: | 26-Dec-2015 |
Category: |
Documents |
Upload: | prudence-walton |
View: | 232 times |
Download: | 3 times |
1 of 25
The EPA 7-Step DQO Process
Step 5 - Define Decision Rules
15 minutes
Presenter: Sebastian Tindall
DQO Training CourseDay 2
Module 14
2 of 25
Step Objective: This step combines Steps
1 - 4 to produce the following major elements to form decision rules:– Parameter of interest
– Unit of decision making
– Action level
– Alternative actions
Step 5: Define Decision Rules
Step 4: Specify Boundaries
Step 2: Identify Decisions
Step 3: Identify Inputs
Step 1: State the Problem
Step 5: Define Decision Rules
Step 6: Specify Error Tolerances
Step 7: Optimize Sample Design
3 of 25
Alternative Actions Step 2
Population of Interest
Step 4
Basis for Defining Action Level
Step 3
Step 5- Define Decision Rules
Develop a Decision Rule
If/Then Decision
Rule Statements
Information IN Actions Information OUT
From Previous Step To Next StepPrincipal Study Question
Step 2
Specify the Parameter of Interest
Confirm the Action Level
Information Required to Resolve Decision Statement
Step 3
Unit of Decision Making
Step 4
4 of 25
Decision Rule
General Format
If theIf the [parameter of interest [parameter of interest (4 elements)(4 elements)]] within thewithin the [unit of decision [unit of decision (3 elements)(3 elements)]] is is >> the the [action level],[action level], then takethen take [alternative action A];[alternative action A]; if notif not [alternative [alternative action B].action B].
5 of 25
Major Elements of a Decision Rule: Population Parameter of Interest
– Population Parameter (ALWAYS UNKNOWN)
– Sample Statistic (Used to represent the Population Parameter)
– Environmental Variable• Chemical/physical attribute in the population• Levels measured (quantity)
Unit of Decision Making (Step 4)– Population (lives with Decision Unit)
– Geographic Area/Volume (Population Spatial Boundary)
– Timeframe (Population Temporal Boundary)
Action Level (Step 3) Alternative Actions (Step 2)
Step 5 Background
6 of 25
Major Elements of a Decision Rule: Parameter of Interest
– Population Parameter
Step 5 Background
– Sample Statistic
– Environmental Variable• Chemical/Physical attribute in the population• Levels measured (Quantity)
Unit of Decision Making– Geographic Area/Volume
– Timeframe
– Population
Action Level Alternative Action(s)
A descriptive measure (such as a true mean, true median, or true proportion) that specifies the characteristic or attribute that the decision maker would like to know about the population. By definition, this will always remain unknown. Note: The purpose of any data collection design is to produce environmental data that can be used as a reasonable estimate of this Population Parameter.
7 of 25
Major Elements of a Decision Rule: Parameter of Interest
– Population Parameter
Step 5 Background
– Sample Statistic
– Environmental Variable• Chemical/Physical attribute in the population• Levels measured (Quantity)
Unit of Decision Making– Geographic Area/Volume
– Timeframe
– Population
Action Level Alternative Action(s)
The sample statistic, e.g., the sample mean, which is obtained from the sampling data, is used to estimate the Population Parameter. It is often a specified upper confidence limit about the sample mean.
8 of 25
Major Elements of a Decision Rule: Parameter of Interest
– Population Parameter
Step 5 Background
– Sample Statistic
– Environmental Variable• Chemical/Physical attribute in the population• Levels measured (Quantity)
Unit of Decision Making– Geographic Area/Volume
– Timeframe
– Population
Action Level Alternative Action(s)
The variable is both the COPC (chemical and physical) and the level measured. Examples of the level are concentration or activity or result.
9 of 25
Major Elements of a Decision Rule: Parameter of Interest
– Population Parameter
Step 5 Background
– Sample Statistic
– Environmental Variable• Chemical/Physical attribute in the population• Levels measured (Quantity)
Unit of Decision Making– Geographic Area/Volume
– Timeframe
– Population
Action Level Alternative Action(s)
The smallest, most appropriate subset (sub-population) for which separatedecisions will be made.
10 of 25
Major Elements of a Decision Rule: Parameter of Interest
– Population Parameter
Step 5 Background
– Sample Statistic
– Environmental Variable• Chemical/Physical attribute in the population• Levels measured (Quantity)
Unit of Decision Making– Geographic Area/Volume
– Timeframe
– Population
Action Level Alternative Action(s)
Spatial Boundary
Temporal Boundary
11 of 25
Major Elements of a Decision Rule: Parameter of Interest
– Population Parameter
Step 5 Background
– Sample Statistic
– Environmental Variable• Chemical/Physical attribute in the population• Levels measured (Quantity)
Unit of Decision Making– Geographic Area/Volume
– Timeframe
– Population
Action Level Alternative Action(s)
The total number of objects (samples of soil or sludge or sediment or air, etc.), that are contained within the spatial unit to be studied
12 of 25
Major Elements of a Decision Rule: Parameter of Interest
– Population Parameter
Step 5 Background
– Sample Statistic
– Environmental Variable• Chemical/Physical attribute in the population• Levels measured (Quantity)
Unit of Decision Making– Geographic Area/Volume
– Timeframe
– Population
Action Level Alternative Action(s)
A measurement threshold value ofthe Population Parameter (e.g., true mean)that provides the criterion for choosing among alternative actions.
13 of 25
Major Elements of a Decision Rule: Parameter of Interest
– Population Parameter
Step 5 Background
– Sample Statistic
– Environmental Variable• Chemical/Physical attribute in the population• Levels measured (Quantity)
Unit of Decision Making– Geographic Area/Volume
– Timeframe
– Population
Action Level Alternative Action(s)
The actions that the decision maker wouldtake depending on the value of the sample statistic which is an estimate of the Population Parameter (e.g., true mean).
14 of 25
Alternative Actions Step 2
Population of Interest
Step 4
Basis for Defining Action Level
Step 3
Develop a Decision Rule
If/Then Decision
Rule Statements
Information IN Actions Information OUT
From Previous Step To Next StepPrincipal Study Question
Step 2
Specify the Parameter of Interest
Confirm the Action Level
Information Required to Resolve Decision Statement
Step 3
Unit of Decision Making
Step 4
The purpose of the data collection design is to produceenvironmental data that can be used as a reasonable estimateof the Population (true) Parameter.
Step 5- Define Decision Rules
15 of 25
Alternative Actions Step 2
Population of Interest
Step 4
Basis for Defining Action Level
Step 3
Develop a Decision Rule
If/Then Decision
Rule Statements
Information IN Actions Information OUT
From Previous Step To Next StepPrincipal Study Question
Step 2
Specify the Parameter of Interest
Confirm the Action Level
Information Required to Resolve Decision Statement
Step 3
Unit of Decision Making
Step 4
If not done in Step 3, specify the numerical value that would cause a person to choose between alternative actions.
Confirm that the action level is greater than the detection/quantitation limits for the potential measurement methods identified in Step 3.
Step 5- Define Decision Rules
16 of 25
Alternative Actions Step 2
Population of Interest
Step 4
Basis for Defining Action Level
Step 3
Develop a Decision Rule
If/Then Decision
Rule Statements
Information IN Actions Information OUT
From Previous Step To Next StepPrincipal Study Question
Step 2
Specify the Parameter of Interest
Confirm the Action Level
Information Required to Resolve Decision Statement
Step 3
Unit of Decision Making
Step 4
Develop a decision rule as an “if…then…” statement that incorporates the parameter of interest, the unit of decision making, the action level, and the action(s) that would result from resolution of the decision.
Step 5- Define Decision Rules
17 of 25
If the If the [true mean [true mean (as estimated by the 95% (as estimated by the 95% UCL of the sample mean)UCL of the sample mean) concentration ofconcentration of cadmiumcadmium]] within within [[the concrete rubblethe concrete rubble within within the perimeter of the site]the perimeter of the site] is is > [80 mg/Kg],[80 mg/Kg], then then [the rubble requires remediation];[the rubble requires remediation]; if not if not [the [the rubble will be left in place].rubble will be left in place].
Remedial Action Decision Rule Example
18 of 25
If the If the [true mean [true mean (as estimated by the 90% (as estimated by the 90% UCL of the sample mean)UCL of the sample mean) concentration ofconcentration of cadmiumcadmium]] within within [[the concrete rubblethe concrete rubble in a in a container truck]container truck] is is > [1 mg/L TCLP],[1 mg/L TCLP], then then [the waste rubble will be considered [the waste rubble will be considered hazardous waste and will be disposed of in a hazardous waste and will be disposed of in a TSD facility];TSD facility]; if not if not [the waste rubble will be [the waste rubble will be disposed of in a municipal landfill].disposed of in a municipal landfill].
Waste Designation Decision Rule Example
19 of 25
Decision Rule Example
If the [ true mean (as estimated by the 95% UCL of the sample mean) concentration of U-238] in the [3”x 3” x 6” samples of surface soil within the perimeter of the backyard to a depth of 6 inches] is > [20 pCi/g], then [dispose of soil in a radiological landfill]; if not, [leave the soil in place].
20 of 25
If the [true mean (as estimated by the one sided 90% UCL of the sample mean) concentration of cadmium] in [the metal turnings within a 55 gallon drum] is > [1 mg/L TLCP], then [the metals turnings will be [the metals turnings will be considered hazardous waste and will be considered hazardous waste and will be disposed of in a TSD facility];disposed of in a TSD facility]; if not if not [the [the metals turnings will be disposed of in a metals turnings will be disposed of in a municipal landfill].municipal landfill].
Decision Rule Example
21 of 25
“Typical” Decision Rules
If the 95% UCL is greater than 10 mg/kg, evaluate what to do next.
If the lead concentration is greater than 10 mg/kg, then dispose of the lead.
22 of 25
If the vadose zone soil moisture content, contaminant concentration profiles, and soil physical properties from the 1454 site exceed or deviate significantly from the conceptual model, an additional evaluation will be performed to assess priority of performing further analysis.
The data gathered from the characterization boring located at the 1454 site combined with historical process data and geophysical logging (high resolution spectral gamma-ray and neutron logging) of existing wells located in the vicinity of the 1455 site can be used to create an analogous model for the 1455 site. If the data collected from characterization of the 1454 site and the geophysical logging data from the existing wells located near the 1455 site support the conceptual model, the analogous unit approach is valid for the 1455 site.
“Typical” Decision Rules
23 of 25
Step 5 - Summary After data are obtained and undergo Data
Quality Assessment, decision rules are used to make the decisions
The decision rules encompass the logic used and include inputs from Steps 1-4 and 6
One may need to perform Step 6 and then update Step 5 to include the statistical decision criteria
24 of 25
Alternative Actions Step 2
Population of Interest
Step 4
Basis for Defining Action Level
Step 3
Develop a Decision Rule
If/Then Decision
Rule Statements
Information IN Actions Information OUT
From Previous Step To Next StepPrincipal Study Question
Step 2
Specify the Parameter of Interest
Confirm the Action Level
Information Required to Resolve Decision Statement
Step 3
Unit of Decision Making
Step 4
Step 5- Define Decision Rules