Part I covered how operations are designed: process types, capacity, location, layout, forecasting and basic inventory. Part II is about running them well: planning the work, scheduling it, controlling quality, removing waste, keeping equipment healthy and managing projects.
1. Planning hierarchy and aggregate planning
Planning moves from the long to the short term:
| Level | Horizon | Output |
|---|---|---|
| Long-range | Years | Capacity, location, process choice |
| Aggregate (sales and operations planning) | 3–18 months | Production, workforce and inventory by month for product families |
| Master production schedule (MPS) | Weeks to months | What to make, how many, when |
| MRP and shop-floor scheduling | Days to weeks | Component orders and job sequences |
Aggregate planning matches capacity to forecast demand with levers: chase (vary workforce, hiring or overtime to follow demand), level (steady output, building stock in slack periods) or mixed, plus subcontracting and back-orders. Cost trade-off: hiring, firing, overtime, inventory holding and shortage. Services apply the same idea to staff rosters.
In the real world: a consumer-durables company builds stock before the festival season (level strategy for part of the year), while a restaurant uses a chase strategy, adding part-time staff on weekends.
2. Material requirements planning (MRP)
MRP computes what to order and when, from the demand for finished goods. It needs three inputs: the master schedule, the bill of materials (BOM) (the product tree of parts and quantities) and inventory records. It works on dependent demand: if you make 100 bikes, you need exactly 200 wheels, which are known from the BOM and not forecast separately.
Steps: explode the BOM, compute gross requirements, subtract on-hand and scheduled receipts to get net requirements, then offset by lead time to find the planned order release date.
Example: to deliver 100 bikes in week 5, with 200 wheels needed, 50 in stock and a lead time of 2 weeks, net requirement = 150 wheels, so order them in week 3. Extensions: MRP II adds capacity, finance and shop-floor control; ERP extends to the whole business.
3. Scheduling and sequencing
Scheduling assigns work to resources and times. Priority rules for sequencing jobs on one machine:
| Rule | Logic | Effect |
|---|---|---|
| FCFS (first come, first served) | Order of arrival | Fair, simple |
| SPT (shortest processing time) | Shortest job first | Minimises average flow time |
| EDD (earliest due date) | Most urgent first | Minimises maximum lateness |
| Critical ratio | Time remaining ÷ work remaining | Balances urgency |
Example: jobs of 5, 2 and 8 hours, in SPT order (2, 5, 8): completion times 2, 7, 15, average 8. In FCFS order (5, 2, 8): 5, 7, 15, average 9. Johnson’s rule sequences jobs on two machines to minimise total time: take the smallest time overall; if it is on machine 1, schedule that job first, if on machine 2, last; repeat. Gantt charts show the schedule. Services use appointment systems, reservations and queue management, and Theory of Constraints (drum-buffer-rope) focuses scheduling on the bottleneck.
4. Quality management
Quality is fitness for use and meeting customer expectations. Dimensions (Garvin): performance, features, reliability, conformance, durability, serviceability, aesthetics and perceived quality. Gurus: Deming (14 points, PDCA, system causes), Juran (quality trilogy, fitness for use), Crosby (“quality is free”, zero defects) and Ishikawa (cause-and-effect diagrams, quality circles).
| Tool or idea | Purpose |
|---|---|
| TQM | Customer focus, continuous improvement, employee involvement, process thinking |
| PDCA / Kaizen | Plan–do–check–act cycles; many small improvements |
| Seven QC tools | Check sheet, histogram, Pareto chart (80/20), fishbone, scatter, control chart, flowchart |
| Six Sigma (DMAIC) | Reduce variation to 3.4 defects per million opportunities; Define, Measure, Analyse, Improve, Control |
| ISO 9001 | Certifiable quality-management standard |
| Statistical process control (SPC) | Control charts watch a process over time |
| Acceptance sampling | Accept or reject a lot from a sample |
| Poka-yoke | Mistake-proofing in design |
Control charts: x̄ and R charts for measurements, p and c charts for defects. Points outside the control limits (mean ± 3σ) or non-random patterns signal a special cause. A process can be in control yet fail the specification, so process capability (Cp, Cpk) compares process spread with spec limits.
In the real world: Toyota’s andon cord lets any worker stop the line to fix a problem at the source, building quality in. Motorola created Six Sigma and GE under Jack Welch made it famous. The Boeing 737 MAX crisis is often cited as a case where schedule and cost pressure and weak safety culture undermined quality. Software teams use the same loop with CI tests, SLOs and post-incident reviews.
5. Lean, JIT and waste
Lean removes waste (muda) and creates flow by pull. The seven wastes: overproduction, waiting, transport, over-processing, inventory, motion and defects (an eighth: unused talent).
| Lean tool | Idea |
|---|---|
| Just-in-time (JIT) | Produce and deliver only what is needed, when needed |
| Kanban | Card signals that pull work or parts |
| 5S | Sort, set in order, shine, standardise, sustain |
| SMED | Quick changeover to allow small batches |
| Heijunka | Level production |
| Value-stream mapping | See all steps; separate value-adding from non-value-adding |
| Jidoka | Automation with human touch; stop on defects |
Benefits: lower stock and lead time, fewer defects. Risks: fragile when suppliers fail, which is why firms add buffers for critical items.
In the real world: Toyota Production System is the origin; Maruti Suzuki’s Indian plants use JIT with supplier parks nearby. Hospitals use lean to cut patient waiting, and software teams apply Kanban boards and limit work in progress.
6. Maintenance and reliability
| Type | Description |
|---|---|
| Breakdown (corrective) | Fix after failure; cheap until it is not |
| Preventive | Scheduled servicing at intervals |
| Predictive (condition-based) | Use sensors and data to service when needed |
| Total productive maintenance (TPM) | Operators share care of equipment; target zero breakdowns |
Overall equipment effectiveness (OEE) = Availability × Performance × Quality. Example: 90% × 95% × 98% ≈ 84%. Reliability is the probability of working for a period; MTBF (mean time between failures) and MTTR (mean time to repair) give availability = MTBF ÷ (MTBF + MTTR). Aircraft engines and wind turbines use predictive maintenance from sensor data.
7. Project management and modern trends
Projects are unique, temporary efforts. CPM uses single time estimates; PERT uses optimistic, most likely and pessimistic times, with expected time (a + 4m + b) ÷ 6. The critical path is the longest path with zero float; delays on it delay the project. Crashing shortens the project by adding resources at least cost per day. Example: activities A (3 days) then B (4 days) in parallel with C (5 days), then D (2 days): paths A-B-D = 9 days and C-D = 7, so the critical path is A-B-D.
Trends: Industry 4.0 (IoT sensors, robotics, digital twins, additive manufacturing), sustainability and circular economy, servitisation (selling outcomes, not products), and resilience after supply shocks.
One-minute recap
| Topic | Key idea | Where you meet it |
|---|---|---|
| Aggregate planning | Chase, level or mixed | Festival stocking, staffing |
| MRP | BOM explosion, net requirements, lead-time offset | Manufacturing, ERP |
| Scheduling | SPT, EDD, Johnson’s rule, bottleneck focus | Job shops, appointments |
| Quality | TQM, Six Sigma, control charts, ISO | Factories, services, DevOps |
| Lean and JIT | Seven wastes, pull, 5S, SMED | Toyota, hospitals, Kanban |
| Maintenance | Preventive to predictive; OEE | Plants, aircraft |
| Projects and trends | CPM and PERT; Industry 4.0 | Launches, smart factories |
Next in MBA Revisited: Marketing Management II.
Written by Claude, Anthropic’s AI assistant, as part of MBA Revisited, a quick refresher of the PGDM subjects one at a time.