allocating information gathering efforts for selection ......stochastic simulation optimization: an...

20
Dennis D. Leber National Institute of Standards and Technology Jeffrey W. Herrmann University of Maryland Science of Test Workshop – Springfield, VA April 4, 2017 Allocating Information Gathering Efforts for Selection Decisions

Upload: others

Post on 13-Jul-2020

2 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Allocating Information Gathering Efforts for Selection ......Stochastic Simulation Optimization: An Optimal Computing Budget Allocation. Hackensack, NJ: World Scientific Publishing

Dennis D. Leber

National Institute of Standards and Technology

Jeffrey W. Herrmann

University of Maryland

Science of Test Workshop – Springfield, VA

April 4, 2017

Allocating Information Gathering

Efforts for Selection Decisions

Page 2: Allocating Information Gathering Efforts for Selection ......Stochastic Simulation Optimization: An Optimal Computing Budget Allocation. Hackensack, NJ: World Scientific Publishing

of 15

Selecting a Radiation Detection System

1

Performance Operational Impact Cost

Pd SNM (e.g., WGPu, HEU)

Pd Industrial (e.g., 137Cs, 57Co)

Pd Medical (e.g., 131I, 201Tl )

Probability of false alarm

Page 3: Allocating Information Gathering Efforts for Selection ......Stochastic Simulation Optimization: An Optimal Computing Budget Allocation. Hackensack, NJ: World Scientific Publishing

of 15

Challenge

2

How do we allocate our limited and fixed budget for

information gathering to maximize the probability

of selecting the true best alternative?

Page 4: Allocating Information Gathering Efforts for Selection ......Stochastic Simulation Optimization: An Optimal Computing Budget Allocation. Hackensack, NJ: World Scientific Publishing

of 15

Selection Decision Terminology

3

Alternatives

a1: β€œtownhouse”

a2 : β€œfarmhouse”

a3 : β€œcountry house”

Attributes (value)

1: House size (mi1)

2: Lot size (mi2)

3: Cost (mi3)

4: Distance to work (mi4)

5: Quality of school (mi5)

Preferences and Decision Model

Mike Mulligan and His Stream Shovel

Virginia Lee Burton (1939)

1 1 1 2 2 2 3 3 3 4 4 4 5 5 5i i i i i iv v v v v m m m m m

Page 5: Allocating Information Gathering Efforts for Selection ......Stochastic Simulation Optimization: An Optimal Computing Budget Allocation. Hackensack, NJ: World Scientific Publishing

of 15

Best Alternative

4

Alternative that provides the largest decision value

Function of:

1. Decision-maker’s preferences

2. True attribute values

Knowledge of decision values is uncertain

Uncertainty in attribute values is a function of amount of information gathered

1 1 1 2 2 2i i i k k ikv v v m m m

1 1 1 2 2 2Λ† Λ† Λ† Λ†i i i k k ikv v v m m m

-------------- Estimate --------------

Page 6: Allocating Information Gathering Efforts for Selection ......Stochastic Simulation Optimization: An Optimal Computing Budget Allocation. Hackensack, NJ: World Scientific Publishing

of 15

Selecting an Alternative

5

Probability that alternative ai has largest decision value: 𝑝𝑖 = 𝑃 πœ‰π‘– > πœ‰π‘Ÿ, βˆ€π‘Ÿ = 1,… ,π‘š

Select as, where 𝑠 = argmax𝑖

𝑝𝑖

Define Probability of Correct Selection: PCSDM = 𝑝𝑠

Page 7: Allocating Information Gathering Efforts for Selection ......Stochastic Simulation Optimization: An Optimal Computing Budget Allocation. Hackensack, NJ: World Scientific Publishing

of 15

Information Gathering Layout

6

Assume equal cost for all observations

B is observation budget

Performance

Measure 1

Performance

Measure 2…

Performance

Measure k

Alternative a1 𝑛11 n12 n1k

Alternative a2 𝑛21 n22 n2k

… … … …

Alternative am π‘›π‘š1 nm2 nmk

Page 8: Allocating Information Gathering Efforts for Selection ......Stochastic Simulation Optimization: An Optimal Computing Budget Allocation. Hackensack, NJ: World Scientific Publishing

of 15

Uniform Allocation

7

Performance

Measure 1

Performance

Measure 2…

Performance

Measure k

Alternative a1 𝑛11 =𝐡

π‘šπ‘˜π‘›12 =

𝐡

π‘šπ‘˜π‘›1π‘˜ =

𝐡

π‘šπ‘˜

Alternative a2 𝑛21 =𝐡

π‘šπ‘˜π‘›22 =

𝐡

π‘šπ‘˜π‘›2π‘˜ =

𝐡

π‘šπ‘˜

… … … …

Alternative am π‘›π‘š1 =

𝐡

π‘šπ‘˜π‘›π‘š2 =

𝐡

π‘šπ‘˜π‘›π‘šπ‘˜ =

𝐡

π‘šπ‘˜

Page 9: Allocating Information Gathering Efforts for Selection ......Stochastic Simulation Optimization: An Optimal Computing Budget Allocation. Hackensack, NJ: World Scientific Publishing

of 15

Proportional Allocation

8

Performance

Measure 1

Performance

Measure 2…

Performance

Measure k

Alternative a1 𝑛11 = πœ†1𝐡

π‘šπ‘›12 = πœ†2

𝐡

π‘šπ‘›1π‘˜ = πœ†π‘˜

𝐡

π‘š

Alternative a2 𝑛21 = πœ†1𝐡

π‘šπ‘›22 = πœ†2

𝐡

π‘šπ‘›2π‘˜ = πœ†π‘˜

𝐡

π‘š

… … … …

Alternative am π‘›π‘š1 =πœ†1

𝐡

π‘šπ‘›π‘š2 =

πœ†2𝐡

π‘šπ‘›π‘šπ‘˜ = πœ†π‘˜

𝐡

π‘š

1 1 1 2 2 2i i i k k ikv v v m m m

Page 10: Allocating Information Gathering Efforts for Selection ......Stochastic Simulation Optimization: An Optimal Computing Budget Allocation. Hackensack, NJ: World Scientific Publishing

of 15

Sequential Information Gathering

9

Page 11: Allocating Information Gathering Efforts for Selection ......Stochastic Simulation Optimization: An Optimal Computing Budget Allocation. Hackensack, NJ: World Scientific Publishing

of 15

Ranking and Selection

10

Statistics

Bechhofer, R. E. (1954). A Single-sample Multiple Decision Procedure for

Ranking Means of Normal Populations with Known Variances. The Annals of

Mathematical Statistics, 25, 16-39.

Bechhofer, R. E., Santer, T. J., & Goldsman, D. M. (1995). Design and Analysis of

Experiments for Statistical Selection, Screening, and Multiple Comparisons. New

York: John Wiley and Sons, Inc.

Computer Simulation Kim, S.-H., & Nelson, B. L. (2006). Selecting the Best System. In S. G.

Henderson, & B. L. Nelson (Eds.), Handbook in Operations Research and Management Science (Vol. 13, pp. 501-534). Oxford: Elsevier.

Chen, C.-H., & Lee, L. H. (2011). Stochastic Simulation Optimization: An Optimal Computing Budget Allocation. Hackensack, NJ: World Scientific Publishing Co. Pte. Ltd.

Page 12: Allocating Information Gathering Efforts for Selection ......Stochastic Simulation Optimization: An Optimal Computing Budget Allocation. Hackensack, NJ: World Scientific Publishing

of 15

Amalgamation

11

𝑛𝑖𝑗 𝑑

Prior knowledge of

attribute values, πœ‡π‘–π‘—β€™s Knowledge of

Decision-maker

preferences, πœ†π‘—Updated knowledge of

attribute values, πœ‡π‘–π‘—β€™s with collected data, 𝑿 𝒕

Prior knowledge of ability

to measure attribute

values, U ΖΈπœ‡π‘–π‘— = πœŽπ‘—β€™s

Update knowledge of ability

to measure attribute values,

πœŽπ‘—β€™s with collected data, 𝑿 𝒕

Page 13: Allocating Information Gathering Efforts for Selection ......Stochastic Simulation Optimization: An Optimal Computing Budget Allocation. Hackensack, NJ: World Scientific Publishing

of 15

Sequential Allocation Procedure

12

Data collected thus far for alternative ai and attribute j: 𝐱𝑖𝑗 𝑑 = π‘₯𝑖𝑗1, … , π‘₯𝑖𝑗𝑛𝑖𝑗 𝑑

Knowledge of value of attribute j for alternative ai: 𝑝 πœ‡π‘–π‘—|𝐱𝑖𝑗 𝑑

Knowledge of decision value for alternative ai: 𝑝 πœ‰π‘–|𝐱𝑖 𝑑

Calculate PCSDM 𝑑 = 𝑃 πœ‰π‘  > πœ‰π‘Ÿ , βˆ€π‘Ÿ = 1,… ,π‘š|𝐗 𝑑

Page 14: Allocating Information Gathering Efforts for Selection ......Stochastic Simulation Optimization: An Optimal Computing Budget Allocation. Hackensack, NJ: World Scientific Publishing

of 15

Allocation Decision

13

Next sample

Posterior predictive distribution

Allocate sample to π‘Žπ‘–, 𝑗 with

largest Expected PCS𝐷𝑀𝑖𝑗

𝑑 + 1

11

ijijDMijn t

x PCS t

1|

ijijijn t

p x t

x

1 11 1 11, , ,

1, ,

1 max , , | , |ij ij ij ij

q r

DM m m iijn t ijn t ijn tq m r q

r m

E PCS t p t x d d p x t dx

X x

Page 15: Allocating Information Gathering Efforts for Selection ......Stochastic Simulation Optimization: An Optimal Computing Budget Allocation. Hackensack, NJ: World Scientific Publishing

of 15

Allocation Procedure Performance

14

Evaluation Experiment:

50,000 decision cases

Concave efficient frontier

π‘š = 5 alternatives

π‘˜ = 2 attributes

100 ≀ πœ‡π‘–π‘— ≀ 200

19 decision models, πœ†1, πœ†2 pairs

𝑣𝑗 πœ‡π‘–π‘— = πœ‡π‘–π‘— Gaussian measurement error (known)

Bayesian prior on attribute values

𝑝 πœ‡π‘–π‘— = 𝑁 150, 352

Experimental budget 𝐡 = 50

Observed frequency of correct selection

Page 16: Allocating Information Gathering Efforts for Selection ......Stochastic Simulation Optimization: An Optimal Computing Budget Allocation. Hackensack, NJ: World Scientific Publishing

of 15

Summary

15

Allocating a fixed experimental budget across multiple attributes and

alternatives in a selection decision where the results of the experimental

evaluations lead to uncertain estimates of the true attribute values

Allocation approaches:

1. Uniform

2. Proportional

3. Sequential

Allocation does impact the probability of selecting the true best alternative

Importance for projects focused on a selection decision to be managed so

that the decision modeling and the experimental planning are done jointly

rather than in isolation

Page 17: Allocating Information Gathering Efforts for Selection ......Stochastic Simulation Optimization: An Optimal Computing Budget Allocation. Hackensack, NJ: World Scientific Publishing

BACKUP

16

Page 18: Allocating Information Gathering Efforts for Selection ......Stochastic Simulation Optimization: An Optimal Computing Budget Allocation. Hackensack, NJ: World Scientific Publishing

of 15

How should we allocate a fixed budget across multiple attributes

and alternatives to maximize the probability of correct selection?

Assumptions and Problem Statement

17

1. Finite and distinct set of alternatives

attributes; true value

Separate and independent attribute measurement processes

with known

2. Decision model is provided

Linear

3. Fixed experimental budget

B sample measurements; cost equivalent

1, , ma a

1 1, ,

k

i i ik j j ijjf v m m m

2k ijm

ijl ij ijlX m

2~ 0,ijl jN 2

j

j ij ijv m m1

1k

jj

Page 19: Allocating Information Gathering Efforts for Selection ......Stochastic Simulation Optimization: An Optimal Computing Budget Allocation. Hackensack, NJ: World Scientific Publishing

of 15

Bayesian Estimation

18

Estimate by estimating each

Decision-maker’s prior knowledge

Data

Posterior distribution on

Posterior distribution on

ijm

2

0 0,ij ijN m i

1, ,ijij ijnx x

ijm

i

2 2 2 2

0 0 0

1 2 2 2 2

0 0

| , , ~ ,ij

j ij ij ij ij j ij

ij ij ijn

j ij ij j ij ij

n xx x N

n n

m m

2 2 2 2

0 0 02

2 2 2 210 0

| ~ ,k k

j ij ij ij ij j ij

i i j j

j i jj ij ij j ij ij

n xN

n n

m

x

Page 20: Allocating Information Gathering Efforts for Selection ......Stochastic Simulation Optimization: An Optimal Computing Budget Allocation. Hackensack, NJ: World Scientific Publishing

of 15

Results

19