Manufacturing + operations¶
-
Prerequisites
Modules 4-6; Wave 1 optimization/metrology -
Exit capability
Translate prototypes into processes with rate, yield, variation, quality, cost, maintenance, supply chain and learning curves. -
Unlocks / transfers to
Robot factories; orbital manufacturing; batteries at scale; biotech production; chip fabs; additive manufacturing; autonomous construction.
Weeks¶
Week 45¶
Spine: MIT 2.008 Spring 2025
Reading: Process physics and design-for-manufacturing lectures; machining/deforming/casting/additive overview
Know: Choose manufacturing processes from geometry/material/rate/quality/cost constraints.
Reconstruct: Regenerate chip/load/energy or material-flow scaling for one process and basic tolerance-stack logic.
Do: Take a Wave-2 component and compare machining, forming/casting and additive routes quantitatively.
Defend: Why is a manufacturable geometry different from an optimally shaped geometry?
Gate: Pass: process selection includes tooling, rate, tolerances, material utilization and inspection.
Source: source
Week 46¶
Spine: MIT 2.008 Spring 2025
Reading: Lectures 11-12 variation/quality/statistical process control
Know: Translate metrology into yield/capability/process control and diagnose variation sources.
Reconstruct: Derive Cp/Cpk intuition and control-limit standard-error scaling.
Do: Simulate process drift and compare inspection-only vs process-control strategies.
Defend: Why can 100% inspection still produce poor quality?
Gate: Pass: distinguish measurement error, common cause, special cause and specification.
Source: source
Week 47¶
Spine: MIT 2.008 Spring 2025
Reading: Lectures 15-19 manufacturing systems, planning, cost, lean, transfer lines
Know: Reason about capacity, bottlenecks, WIP, cycle time, utilization, flow and production economics.
Reconstruct: Derive Little's Law and bottleneck throughput bound.
Do: Scale a prior prototype to 10,000 units/year: routing, machines, staffing/automation, WIP, downtime and unit cost.
Defend: Why does maximizing machine utilization often hurt system throughput?
Gate: Pass: line design includes bottleneck, variability and recovery, not average cycle time only.
Source: source
Week 48¶
Spine: MIT 2.008 + Factory Physics/Groover reference
Reading: Integrated production-system studio
Know: Integrate process physics, quality, maintenance, supply chain, learning curve and capital deployment.
Reconstruct: Derive yield multiplication across serial process steps and simple learning-curve relation.
Do: Create manufacturing plan for a battery module, robot actuator, biosensor or spacecraft component with supplier and maintenance risks.
Defend: Where should redundancy live: product, process, supplier or inventory?
Gate: Module defense: credible 10k-unit plan with rate/cost/yield/quality/maintenance/supply evidence.
Source: source
Exit gate¶
Closed-book: 120 min: process selection, yield/capability, Little's Law, bottlenecks, cost, maintenance, learning curves.
Novel problem: Scale one previous prototype to 10,000 units/year with credible routing and quality system.
Artifact: Manufacturing plan with rate, WIP, yield, downtime, staffing/automation, tooling, suppliers and unit cost.
Defend: Defend bottleneck, variation, maintenance, supplier risk and make/buy choice.
Pass criterion: Pass if throughput/cost/yield numbers reconcile and recovery from disruption is modeled.
Transfer problems¶
Try these before consulting solutions or asking for the complete answer.
-
Process selection: Choose process for 10k parts/year and justify material, geometry, tolerance, tooling and cost.
-
Yield: Compute total line yield from serial step yields and identify highest-leverage improvement.
-
Capability: Calculate Cp/Cpk and explain what they do not guarantee.
-
SPC: Simulate drift and design detection rule balancing false alarms/delay.
-
Little's Law: Relate throughput, WIP and cycle time in a production line.
-
Bottleneck: Identify throughput bottleneck and quantify effect of adding capacity elsewhere.
-
Downtime: Model OEE/availability impact of MTBF/MTTR changes.
-
Learning curve: Estimate unit labor/cost after cumulative production doubles several times.
-
Supply chain: Compare dual-source, safety stock and redesign strategies for a critical part.
-
10k plan: Produce routing, takt/rate, machines, shifts, QA, maintenance and unit-cost model for one prior build.
Textbooks¶
See the five-book resource page.