1 / 25

IE 419 Work Design: Productivity and Safety Dr. Andris Freivalds Class #17

IE 419 Work Design: Productivity and Safety Dr. Andris Freivalds Class #17. Ch. 14 – Work Sampling. Work Measurement – to establish a standard time for a given job: Time study (IE 327) Predetermined time systems (IE 419) MTM-2 MOST Standard data (IE 327)

amela-harrell
Download Presentation

IE 419 Work Design: Productivity and Safety Dr. Andris Freivalds Class #17

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. IE 419 Work Design:Productivity and Safety Dr. Andris Freivalds Class #17 IE 419

  2. Ch. 14 – Work Sampling • Work Measurement – to establish a standard time for a given job: • Time study (IE 327) • Predetermined time systems (IE 419) • MTM-2 • MOST • Standard data (IE 327) • Work Sampling – method of find % occurrence of an activity by statistical sampling and random observations IE 419

  3. Work Sampling • Can not continually observe an event • At a given instant, the event either happens or does not • Probability of such binary event happening m out of n times, is determined by the binomial distribution: P(m) = n! pm (1-p)m-n m! (n-m)! n = # obs, m = # events, p = prob of event, q=1-p IE 419

  4. Prob of 0,1,2,3,4 heads in 4 tosses • n! pm (1-p)n-m / m! (n-m)! IE 419

  5. Normal Approximation As n → ∞, binomial → normal, mean = p, std. dev. s = √pq/n IE 419

  6. # Observations Needed • 95% fall in interval p ±ℓ (limit of error) • Where ℓ =1.96s = 1.96 √pq/n • ℓ2 = 3.84 pq/n → n = 3.84pq/ℓ2 • Ex 1: Machines down 25% of time. How many observations needed for 6% error? • Ex 2: Only collect n=100. What is error? IE 419

  7. Frequency and Timing of Observations • Use random number table • 4 digits: #1=day, #2 =hr, #3-4 = min. • Discard outside range • Arrange in numerical order • Or use DesignTools! IE 419

  8. Steps in Work Sampling • Define problem, event to study • Pilot study, for initial p • Determine l, then n • Determine random observations • Design recording form • Carry out study IE 419

  9. Recording Form - Simple IE 419

  10. Recording Form - More IE 419

  11. Recording Form - Detailed IE 419

  12. Problem - Correlated Observations • If several workers observed simultaneously • Individual readings are not independent • Need to correct standard deviation (s) and interval p +ℓ or p +1.96s • s2 = ∑yj2/nj – np2 n(m-1) • Where: n = total #obs, m = # grouped obs • Where: nj = #workers at jth obs, yj = #workers “idle” at jth observation IE 419

  13. Correlated Observations - Ex Standard (but incorrect) approach: ℓ2 = 3.84 pq/n = 95% confident that interval:28 ± 12.4% (Incorrect!) IE 419

  14. Correlated Observations – Ex #3 s2 = [∑yj2/nj–np2]/n/(m-1) = s = 0.086 and p=14/50 = 0.28 95% confident that interval: 28 ± 1.96(8.6) or 28 ± 16.4% (correct) IE 419

  15. Applications of Work Sampling • Calculation of Standard Time • Finding Worker or Machine Utilization • Application to Service Industry • Self Observation IE 419

  16. Calculating a Standard Time Ex #4 (Press operator for 8-hr shift) Observed time (OT) = Tx(ni/n) / #P Normal time (NT) = OTx R/100 Standard time (ST) = NT(1+Allow) IE 419

  17. Calculating a Standard Time (From Work Sampling) • Observed time (OT) = Tx(ni/n) / #P • Normal time (NT) = OTxR/100 • Standard time (ST) = NT(1+Allow) IE 419

  18. Ex #5 - Finding Worker Utilization IE 419

  19. Ex #6 -Application to Service IndustryWashington State Dept. of Social and Health Services - 1 • Consultant to Sterling Associates in 1999 • Allocation of workload to social workers • Buzzers given to 304 workers • Over 2 months found: • 91,371 observations recorded • 2,224 cases processed, by AU (μ =1 hr/AU) • % idle (4%) very low • Accuracy high, ℓ = 0.4% IE 419

  20. Washington DSHS - #2 Large disparity between AU and Time Redistribution of staffing assignments. IE 419

  21. Self Observation - 1 • Find “typical” day of college student • % of 24-hour day for different activities: • sleeping (record this directly) • classes (sample this and others) • studying • sports/leisure activities (TV, internet) • eating • transportation, going to-from classes • miscellaneous IE 419

  22. Self Observation - 2 • Key issue: • Random reminder • Use alarm • Or use PDA • Palm Zire • OS 4.1 • QuikSamp IE 419

  23. Flowchart of QuikSamp START Randomizing # of Observ. Limit of Error Sample Days and Time Sorting & Display Work Sampling # of Operator & Work Elements Work sampling RESULT END IE 419

  24. Potential Sources of Error • Sampling error – statistics, ↑n • Bias errors: • Observer bias – expecting outcomes • Operator bias – misleads analyst • In-phase bias – regular sampling bad, random • Representativeness error – sampling during nonrepresentative time, e.g. holiday IE 419

  25. Work Sampling vs. Time Study IE 419

More Related