Study Guides (238,342)
Canada (115,087)
Sociology (440)
SOC 202 (61)

Chapter 10.docx

4 Pages
Unlock Document

Ryerson University
SOC 202
Carmen Schifellite

Chapter 10: Statistical Quality Control LO 1: Introduction best companies = emphasize designing quality into process (contin improvement, six sigma) → reduce need for inspection/test goal: level of quality can avoid inspection test and process control statistical quality control - use of stat techniques and sampling in monitoring and testing of quality of goods and services - acceptance sampling - relies primarily on inspection/tests of prev produced items - stat process control - stat quality control that occurs during production - eval quality of products and meet expectations of customers inspection - appraisal of good/service against a std - verify not contain more than specified % of defective goods → physically examine Statistical process control planning process 1. define the quality characteristics - define in detail what to be controlled o diff characteristics req diff approaches for control purposes o std → eval measurement 2. for each characteristics a. determine quality control point  each inspection adds cost to product  HACCP provide some guidelines 1. at beg of process - little sense in paying for goods not meet quality std and spending time and effort 2. at end of process - customer satif and comp image and repairing/replacing products costly 3. at operation where characteristic of interest to customers is first determined - b4 costly/irreversible b. plan how inspection done  tech and need engineer knowledge how much to inspect  low cost, high vol = little inspection b/c: 1) cost associated w/ passing defective items low, 2) processes that produce these items highly reliable so defects rare  amt inspection needed governed by cost of inspection and expected cost of passing defective item s  inspection activities ↑ → inspection cost ↑→ cost of passing defective ↓  goal: min sum of those 2 costs  operation w/ a high prop of human involvement → more inspection → more reliable  freq of inspection depends on rate process may go out of control centralized vs. onsite inspection  on site - measuring dimension  adv: specialized equip, skilled quality control inspectors, more favourable test envir  disadv: time, interrupted c. plan corrective action  process out of control → uncover cause  off specs → tested → reworked/scrapped  process monitored → verify prob elim LO 2: Statistical process control statistical process control - stat eval of product in production process - periodic samples compare w/ predetermined limits - results outside lime → stop process → corrective action - w/in limits → contin types of variations and sampling distributions - random variation - natural variation in output of a process created by countless minor factors o older machine → higher degree random variation → f/ worn parts o new machine incorp design improvements → lessen variability in output - assignable variation - non-random variability in process output; variation cause can be identified o main sources identified and elim o defective material, human error o sampling distrib - variability of sample stat, theoretical distrib of val of stat for all possible sample of given size f/ process o process variability - amt arrange on graph, freq distrib o averaging - sampling distrib of sample mean exhibits less variability than process distrib → both distrib have same mean o central limit theorem - sample distrib normal even if pop/process not o sampling distrib - whether process shifted (assignable cause) → not shift = sample mean b/ ±2 (95.5% prob) or ±3 (99.7% prob) std dev control charts - time ordered plot of sample stat w/ limits - distinguish b/ random and assignable variation - process mean - sample mean fall w/in ±3 std dev of mean → set limit at ±3 std dev - assignable variation - limit reflects shift in process - control limits - dividing line for val of sample stat b/ concluding no process shift and process shift → random and assignable variations o deviation f/ long standing know process mean so large process must be shifted o specification limits - based on desired characteristic of product → if process val w/ spec limits - type 1 error - conclu process has shifted (assignable variation present) when has not (only random variation present) o α risk - sum of prob in 2 tails o using wider limits ±3 sigma limits → ↓ prob of type 1 error → but difficult to detect
More Less

Related notes for SOC 202

Log In


Don't have an account?

Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.