Work study: time and motion
Observed time is the directly measured cycle time. Normal time = observed time x performance rating. Standard time = normal time x (1 + allowance fraction), where allowances account for fatigue, personal needs, and delays. Methods study precedes time study: improve the method before measuring it.
Check yourself: Have you applied both the performance rating and allowances?
Inventory: EOQ model
Economic Order Quantity: EOQ = √(2DC_o/C_c), where D is annual demand, C_o is ordering cost per order, and C_c is carrying cost per unit per year. Total cost is minimized at EOQ, where annual ordering cost equals annual carrying cost. The model assumes constant demand, instantaneous replenishment, and no stockouts.
Check yourself: Are the assumptions (constant demand, no shortages) valid for this problem?
Scheduling and sequencing
For n jobs on 2 machines, Johnson's rule minimizes makespan: assign jobs with smaller processing time on machine 1 to the front, and jobs with smaller processing time on machine 2 to the back. PERT uses three time estimates (optimistic a, most likely m, pessimistic b): expected time = (a + 4m + b)/6, variance = ((b - a)/6)².
Check yourself: Have you identified the critical path before computing project duration?
Linear programming
A linear program maximizes or minimizes a linear objective subject to linear inequality constraints. The feasible region is a convex polygon (in 2D). The optimal solution lies at a vertex of the feasible region. The simplex method moves between adjacent vertices. Dual variables give shadow prices for constraints.
Check yourself: Have you checked all vertices of the feasible region for the optimum?
Quality control: control charts
X-bar and R charts monitor process mean and variability. Control limits are set at ±3 standard deviations from the process mean (for X-bar chart). A process is 'in control' when all points fall within control limits with no non-random patterns. Cp = (USL - LSL)/(6σ) measures process capability; Cp ≥ 1.33 is commonly targeted.
Check yourself: Are you distinguishing specification limits (design) from control limits (process)?