Chapter 5 Strategic Capacity Planning Phases of Quality Control: Acceptance Sampling; Statistical Moving Range Control Chart is for moving range, i.e., the
Capacity is the upper limit on the workload that an operating
Process Control; Continuous Improvement and Six Sigma difference between consecutive observations, used to monitor
unit can handle. Capacity is also measured as maximum Inspection Plan: How Much/How Often if inspection the dispersion or spread when n = 1. UCL/LCL =
production rate. Alternatively, a major input is used, eg. size. activities increase, the cost of inspection increases, but the
Importance of Long-term Capacity: Impacts ability to meet cost of defects decreases. Where/When: Raw materials and
Control Charts for Attributes
future demands, Affects operating costs, Major determinant of purchased parts, Before a costly operation; Before an P-chart: is a control chart for sample proportion of defectives,
initial costs, Involves long-term commitment, Affects irreversible process; Before a covering process, Finished it is used to monitor the proportion of defective items
competitiveness. products. Rule of thumb: operations with a high proportion of
generated by a process. (uses normal distribution):
Measuring Capacity human involvement necessitate more inspection than UCL = , LCL = , where:
Design capacity: maximum obtainable output under ideal mechanical operations. Centralized (in-lab) vs. On-site:
conditions. Effective capacity: Maximum capacity given Onsite: immoveable product; simple or handheld measurement
delays, product mix, scheduling difficulties, and other realities. equipment; automated inspection, In-lab: specialized
Efficiency is the ratio of actual output to effective capacity equipment; skilled quality control inspectors; more favourable Ex. Total # of observations = # of observations x # of samples.
test environment. z = # of sigmas of control needed.
C-chart: is a control chart for sample number of defects per
Utilization is the ratio of actual output to design capacity or Statistical Process Control is concerned with statistical unit product, used to monitor the number of defects per unit.
used time over available time. evaluation of the product in the production process. A sample (uses poisson distribution because it can be continuous):
is obtained and compared to the predetermined limits and
UCL = c , LCL = c , where:
either takes corrective action or continues.Types of Variation:
Random variation: Natural variations in the output of process, and .
Factors Influencing Effective Capacity created by countless minor factors. Assignable variation: Non- **When the LCL is negative, it is set to 0.
Facilities: Floor space, layout. Products or services: Limited
menu in a restaurant. Human: Training, skills and experience. random variability in process output; a variation whose source Ex.
Planning and Operational: No of shifts per day, inventory, can be identified and eliminated. Main task of SPC is to
distinguish assignable from random variation. A process should be investigated when a sample statistic
quality control. External: Pollution standards, paper work
Capacity Planning Process falls outside the control limits. There is a sustained upward or
1. Forecast demand one to five years ahead Control Chart basis for the chart is sampling distributions and downward trend. There is a cyclical wave pattern. There is a
the upper and lower control limits define the range of bias of too many observations on one side of the centre line.
2. Determine capacity requirements
3. Measure the capacity now and decide how to bridge the acceptable variation. Control Limits are the dividing lines for There is a level shift. There is too much dispersion.
gap: a) Generate feasible alternatives, b) Evaluate alternatives the value of sample statistic between concluding no process
shift and a process shift. A statistic that falls between the UCL Process Capability: Design specification (DS) is a range of
considering economic and non economic aspects; c) Choose
the best alternative and implement it and the LCL suggest random variation whereas one that falls acceptable values established by engineering design or
Calculating Capacity Requirements outside those limits suggests assignable variation (only a small customer requirements. Process variability (PV) is the actual
percentage of sample means fall more than 2 or 3 standard variability in a process for a product. Process capability is the
deviations from the process mean.
ability of a process to meet the design specification; called
Developing Capacity Alternatives Type I Error is where the error concludes that a process has capability analysis. **Process is capable when PV is less than
shifted when it has not. Wider limits would reduce this error but variation allowed by DS.
1. Design flexibility into the system would also increase the chance of missing assignable errors if Capability Analysis: Capability analysis determines whether
2. Differentiate between new and mature products
3. Take a big picture approach to capacity changes they are present. Type II error is where the error concludes a the process output falls within the design specification. If it is
4. Prepare to deal with capacity chunks process has not shifted when it has. within the design specification it is said to be capable. If not,
management must take corrective action: redesign process or
5. Attempt to smooth out capacity requirements
6. Use capacity cushion Sample Mean and Range Control Charts reduce variability, use alternative process, use 100-percent
7. Identify the optimal operating level Sample Mean Control Chart is used to monitor the process inspection, or relax design specification.
mean. The control limit can be determined in two ways: Process variability is measured 3 standard deviations from
Minimum cost & optimal operating rate are functions of size of
production unit. 1.Using standard deviation of the process: the mean. To determine whether the process is capable,
Economies of Scale UCL = or = , where is the grand mean, z compare the 3 STDs value to acceptable range of variation.
Economies of scale: Fixed costs (facilities, equipment, is the Standard Normal Deviate, and is the process
management) spread out over more units. Volume purchase Process Capability Ratio
discounts. Diseconomies of scale: Worker fatigue, equipment deviation, LCL = or = .
breakdown, less room for error, difficulties in coordination . Ex. Obtain a 3 sigma control limit for sample mean. Std dev. is
Evaluating Alternatives 0.2