Production Planning: Bottlenecks and Throughput (Video Notes)

Context and Bottleneck

• Transcript discusses a production scenario with processes A, B, and C (referred to as process one, two, and three).
• There is a total number of hours available and a time requirement per unit for products X and Y (the exact per-unit times are referenced but not explicitly quantified in this excerpt).
• Process two is identified as the bottleneck (the constraint).
• The goal is to understand how to improve the system given the bottleneck and to rank products by profitability or efficiency with respect to the bottleneck constraint.

Improvement idea: Increase efficiency of maintenance for process two

• Proposed action: Increase the efficiency of the maintenance routine for process two.
• Rationale in the transcript:
  • Process two is the bottleneck; improving its efficiency frees up more hours.
  • Two main purposes of maintenance efficiency improvements:
  • Performance measurement: to evaluate how well the system performs under current constraints.
  • Capacity effect: the bottleneck should be running flat out; reducing downtime means the bottleneck can process more units per hour.
• Throughput capture concept:
  • Throughput capture ratio should be greater than one to indicate profitability when operating costs are considered.
  • If the ratio exceeds operating costs, the production yields a profit.
• Per-product perspective:
  • Because factory-hour costs are not identical across products, the return on factory hour ratio is computed per product.
  • The ratio differs by product, so priority should depend on the per-product ratio rather than a single factory-hour figure.

Throughput and profitability metrics

• Throughput per unit (for the discussed product) is stated as 42 (units of throughput per unit time, exact units not clarified in transcript).
• Return on factory hour (ROFH) is implied as a per-product metric, used to rank products by profitability per unit of constraint time.
• Key takeaway:
  • Higher ROFH per product indicates a better candidate for prioritization under bottleneck constraints.
  • The ranking should be done per product, not just using a single factory-hour figure.

Constraint hours and capacity

• Industry-wide constraint is the bottleneck time on the bottleneck process (process two).
• Availability used in the example:
  • There are 10 machines, each can be used for up to 40 hours per week.
  • Total available hours: $H = 10 \times 40 = 400 \text{ hours}.$
  • Convert to minutes for the constraint: $M = 400 \text{ hours} \times 60 = 24{,}000 \text{ minutes}.$
• Throughput target and bottleneck time per unit:
  • Widget (the prioritized product) requires 5 minutes of constraint time per unit (as stated later in the transcript).
  • If we consider the widget, and the maximum demand is 25,000 units, the total time to meet full demand would be: $T_{demand} = 25{,}000 \times 5 = 125{,}000 \text{ minutes}.$
• Capacity constraint implications:
  • Given the total available minutes $M = 24{,}000$, the maximum units that could be produced (if only constrained by time) would be:
  • $U_{\,\max} = \left\lfloor \frac{M}{5} \right\rfloor = \left\lfloor \frac{24{,}000}{5} \right\rfloor = 4{,}800.$
  • Therefore, you should not plan to produce more than the demand, i.e., at most 25,000 units, but the capacity under the bottleneck is 4,800 units due to time limits.
• In the transcript, a throughput figure of 2,500 units is mentioned as the resulting throughput under the plan, but the math above suggests a capacity of 4,800 units given the hours. This discrepancy is noted as a point to consider when reconciling the example numbers with the described constraint.

Production ranking and plan (per the transcript)

• Ranking rule used in the single-factory scenario:
  • Higher return on factory hour (ROFH) is better; rank products from highest to lowest ROFH.
  • The widget (referred to as the rightmost item, often spelled as "Bidget" due to transcript noise) ranks number one.
• Priority order:
  • 1st (highest ROFH): Widget (the product on the far right in the ranking shown in the transcript).
  • Other products (A, B, C, etc.) would follow based on their ROFH, but only the widget is explicitly identified as first.
• Production planning rule described:
  • For a single factory, you can determine the order of production by ranking on ROFH or on a time window (e.g., per 30 minutes).
  • The transcript mentions the possibility of a throughput-counting approach as