Behavioral Goal Setting Models g for Operations Management Workshop Stochastic models for warehousing systems October 29, 2009 José Antonio Larco José Antonio Larco , Kees Jan Roodbergen, René de Koster, Jan Dul Rotterdam School of Management Erasmus University Rotterdam y Contact: jlarco@rsm.nl
Introduction Propositions Experiment Results Conclusions Goals are interesting for OM… • Current assumptions of OM models: p – People are predictable, work in a stationary way and are unaffected by external factors (Boudreau, 2003) • Challenging goals have a positive effect on performance • Challenging goals have a positive effect on performance – Meta-analysis 8-16 % performance increase over “do your best”” strategies; (Locke and Latham, 1990) – Well studied: > 239 lab experiments, > 156 field studies (Locke Well studied: 239 lab experiments, 156 field studies (Locke and Latham, 1990)
Introduction Propositions Experiment Results Conclusions Two main questions for OM 600 ormance s(D) 1. How is performance related 500 to goal difficulty? 400 ? – Linear? (Locke and Latham, 1968) Li ? (L mmulative perfo 300 300 k d L th 1968) – Levels-off? ( Locke & Latham , 1982) 200 – Decreases? (See et al, 2006) 100 – – Effects of varying skill level? Effects of varying skill level? Cum 0 0 0 0,2 0,4 0,6 0,8 1 Goal difficulty (probability of success) 2. How do workers regulate 9 8 their work pace? their work pace? eed (picks/min) Goal achieved ? 7 – Acceleration towards goal 6 (Hull, 1932) or deadline? 5 – – A steady state pattern or A steady state pattern or 4 Workspe 3 irregular? 2 Deadline – Effect of varying goals & skill 1 level? e e 0 0 2 4 6 8 Time (min) All this in OM contexts where workers have a fixed time to work.
Introduction Propositions Experiment Results Conclusions Two-fold approach pp 1. Proposition generation: workers as decision makers – Objective : maximize utility/preference Objective : maximize utility/preference • Utility derived from work pace itself • Utility derived from evaluation w.r.t goal – Decision: what work-pace to select? (effort to exert) – Behavioral Economic decision models: • M Myopic: Individuals focus only on the “near future”. i I di id l f l th “ f t ” • Planner: Individuals take into account the utility for the whole period. – Derivation of properties from model 2. Propositions Testing: experimental setting – Total performance p – Work pace measurement
Introduction Propositions Experiment Results Conclusions Scope • Work is repetitive; i.e. work content known; cycle times short • Workers are experienced • Feedback is provided . • • Goal G (units processed) to be achieved before deadline D Goal G (units processed) to be achieved before deadline D • Target G serves as reference for evaluating performance • Workers committed to the goal (Locke and Latham, 1990) Cumulative work s(t) , work pace, ṡ =ds/dt Cumulative work s(t) , work pace, ṡ ds/dt •
Introduction Propositions Experiment Results Conclusions Work pace preference (Yerkes-Dodson Law (1908) Work pace preference (Y k D d L (1908) • Relates (Hancock & Utility rate rhythm function R( ṡ ) Warm,1989) : , ) y y ( ) 3 – Stressor Natural speed 2 – Adaptability/desirability ute) 1 lity units/minu 0 15 20 25 30 35 • Defines: -1 -2 – Maximum desirability Maximum desirability Utility rate (uti -3 – Range of tolerance -4 -5 U • Properties: -6 Work speed (picks/minute) ṡ =ds/dt – Convex function
Introduction Propositions Experiment Results Conclusions Goal induced preference Goal induced preference (Kahneman & Tversky 1979) • Properties: Progress utility curve P(s) Progress utility curve P(s) – Strictly increasing St i tl i i 0,3 Target T ≥ ∈ ' ( ( )) 0 ; [ 0 , ] P s t t D 0,2 P(s) 0,1 tility units) P – Loss aversion 0 + δ < − δ δ > 0 100 200 300 400 500 ( ) ( ); 0 ; P G P G -0,1 -0,2 Utility (ut – Diminishing sensitivity -0,3 > < ' ' ( ( )) 0 ; P s t s G -0,4 < > ' ' ( ( ( ( )) )) 0 ; ; P s t s G -0,5 -0 5 -0,6 Total work (picks) s(t) • Usage in goal theory Heath et al.,1999; Steel & Koning, 2006 and Wu, et al. 2008
Introduction Propositions Experiment Results Conclusions 1 Myopic Conjecture 1. Myopic Conjecture + max ( ) ( , ) & & R s Q s s & s Initial Initial conditions conditions : : = ( 0 ) 0 & s Progress utility rate (apply chain rule) - unit consistenc y : ′ = ( ( , , ) ) ( ( ) ) & & Q Q s s P s s First order conditions : ′ & ′ = − ( ) ( ) R s P s
Introduction Propositions Experiment Results Conclusions 1. Myopic Conjecture (Work Pace Propositions) 1 M i C j t (W k P P iti ) >Consistent with goal gradient hypothesis (Hull, 1932) 40 35 te ) icks/minut 30 25 k speed (pi 20 15 Target 100 Work Ma im m speed achie ed at target Maximum speed achieved at target 10 Target 150 Target 200 5 Target 250 0 0 2 4 6 8 10 Time (minutes)
Introduction Propositions Experiment Results Conclusions 2 Planning Conjecture 2. Planning Conjecture ∫ D ′ + max ( ( ) ( ) ) & R s P s dt & s 0 Boundary condition : = ( 0 ) 0 & s D ∫ ′ = − Recognize that ( ) ( ( )) ( ( 0 )) & P s s dt P s D P s <Independent of work pace! 0 Applying pp y g euler formula : = ( ) s t ct <Constant work pace! = using the fact that ( ) constant : & s t = − ( ) ( ) & & R' s D P' D s
Introduction Propositions Experiment Results Conclusions 2 Planning Conjecture (Work Pace Propositions) 2. Planning Conjecture (Work Pace Propositions) >Consistent with Planning Conjecture (Parkinson, 1955) 28,0 minute) 27,5 dt (picks/m 27,0 26,5 Target 400 peed ds/d Target 350 Target 350 Target 300 26,0 Target 200 Target 150 Natural pace p 25,5 5,5 Work s 25,0 24,5 0 1 2 3 4 5 6 7 8 9 10 Time t (minutes)
Introduction Propositions Experiment Results Conclusions Goals and Performance: Contrasting propositions Goals and Performance: Contrasting propositions 295 290 285 (picks) 280 275 rformance 270 265 Per 260 260 Myopic 255 Planner 250 245 0 50 100 150 200 250 300 350 400 450 Goal (picks)
Introduction Propositions Experiment Results Conclusions E periment Design Experiment Design • Simple order picking task, short cycled (<10 sec) • Control for learning effects (previous picking rounds) C t l f l i ff t ( i i ki d ) • Within subject design (3x“4”): – Pilot study “Do your best” control (n=36 subjects) Pil t t d “D b t” t l ( 36 bj t ) – 3 randomized goal levels ( 10, 50, 90 th percentile ) per subject (n=81 subjects) ( j ) – Skill proxy: Average work pace of 10 best work pace used - 4 categories constructed – Credit assigned to all subjects regardless of their performance C dit i d t ll bj t dl f th i f • Process view : work pace measured using time stamps
Introduction Propositions Experiment Results Discussion Experiment : Task description
Introduction Propositions Experiment Results Conclusions Results 1: Goal difficulty-performance esu ts Goa d cu ty pe o a ce > Level-off (Locke and Latham (1982)) (Goal/skill Interaction F (1,77)=7.4; p<0.01) > Consistent with planner results 80 of picks) ∆ = 14.3% 75 ce (number 70 ∆ = 11.9% Skill Level 1 ∆ = 12.6% 65 Skill Level 2 l performanc Skill Level 3 60 ∆ = 7.1% Skill Level 4 55 Tota 50 47 59 66 Goal Level (picks) Goal Level (picks)
Introduction Propositions Experiment Results Discussion Results 2: Work speed-stationary behavior 10 2. Work pace o pace 9,5 selection 1. Stable pattern cks/min 9 3. Acceleration 8,5 8 5 towards deadline towards deadline k pace pic 8 7,5 Work Goal 59 picks 7 Goal 66 picks 6,5 Goal 47 picks 6 6 0 1 2 3 4 5 6 7 8 Time (min)
Recommend
More recommend
