Accounting › Section W · Short-term decision making › Topic W.6
Decisions under constrained resources
When there are more orders than machine hours, the product with the biggest margin per unit is not necessarily the one to make first. What counts is how much each product earns from every hour of the resource that is running out.
When a resource runs out
Every business has limits: machine hours, skilled staff, oven space, shelf space, a consultant's billable time. Most of the time demand, not capacity, sets the level of activity. But when customers want more than the business can supply, one resource usually runs out first. That resource is the constraint: whatever is in shortest supply caps what the whole organisation can produce, and the point in a process where work piles up waiting for it is called a bottleneck. A firm making several products then has to decide which orders to fill and which to turn away. This is a short-term decision: over a few months, the capacity is fixed.
Why margin per unit can mislead
With capacity fixed, fixed costs will be the same whatever mix of products is made, so they are irrelevant, and the goal becomes the largest possible total contribution margin. The unit contribution margin from "Contribution margin", selling price less variable cost per unit, is not enough to rank products, because products use different amounts of the scarce resource. A product earning $150 a unit that takes four hours of a bottleneck machine earns less from those four hours than a product earning $40 a unit that takes half an hour, made eight times over. The scarce hours, not the units, are what must be rationed.
The rule: margin per unit of the constraint
Divide each product's unit contribution margin by the amount of the constrained resource it uses:
Rank the products from highest to lowest on this measure. Then fill demand in that order: give the top-ranked product all the hours its demand requires, then the next, until the resource is used up. The last product in line receives only the hours that remain. The same method works for any constraint, such as contribution margin per kilogram of a scarce material or per square metre of display space.
Worked example: ranking at Kittering Joinery
Kittering Joinery, a fictional workshop, has 2,400 hours a month on its CNC router, which every product needs. Bookshelves sell for $180 with variable costs of $108 and take 1.5 router hours; monthly demand is 600. Desks sell for $420, with variable costs of $270, and take 4 hours; demand is 300. Side tables sell for $95, with variable costs of $55, and take 0.5 hours; demand is 1,200. 1. Unit contribution margins: bookshelf $72, desk $150, side table $40. 2. Hours needed: 900 + 1,200 + 600 = 2,700, more than 2,400, so the router is the constraint. 3. CM per router hour: bookshelf $72 ÷ 1.5 = $48; desk $150 ÷ 4 = $37.50; side table $40 ÷ 0.5 = $80. 4. Ranking: side tables, then bookshelves, then desks, the reverse of the ranking by unit margin.
Worked example: building the production plan
Kittering now allocates its 2,400 hours in ranking order. 1. Side tables: all 1,200, using 600 hours. 2. Bookshelves: all 600, using 900 hours; 900 hours remain. 3. Desks: 900 ÷ 4 = 225 of the 300 demanded. 4. Total contribution margin:
5. Compare a plan that favours the biggest unit margin: 300 desks (1,200 hours), 600 bookshelves (900 hours) and 600 tables with the last 300 hours give $45,000 + $43,200 + $24,000 = $112,200. 6. Ranking by margin per hour earns $12,750 more a month from the same router, with fixed costs unchanged.
What an extra hour is worth
The ranking also shows the value of more capacity. An extra hour on the constraint would be used for the best product whose demand is not yet met, so it is worth that product's contribution margin per hour. At Kittering, 75 desks are unmet, so an extra router hour is worth $37.50, and the firm should pay up to that much in extra costs, such as an overtime premium, for each additional hour. Suppose overtime costs $22 an hour beyond the variable costs already counted. Making the 75 desks needs 300 hours, earning 75 × $150 = $11,250 for $6,600 of extra cost, a gain of $4,650, or $37.50 − $22 = $15.50 an hour. Beyond 300 hours all demand is met, so further hours are worth nothing until demand grows.
Relaxing the constraint
Ranking makes the best of the constraint; the next step is to loosen it, which is often called relaxing or elevating the constraint. Options include running overtime or an extra shift, moving staff from non-bottleneck work, buying or hiring more of the scarce machine, outsourcing part of the bottleneck work, cutting setup time between batches, and inspecting parts before they reach the bottleneck so its time is not wasted on defective work. Extra capacity anywhere else adds nothing, because output is set by the bottleneck. This focus on managing the bottleneck is central to the approach known as the theory of constraints.
Qualitative factors and limits
The ranking is a starting point, not the final plan. Customers who buy desks may expect matching bookshelves, so short-changing one product can cost sales of another. A key customer or a contract may require a minimum quantity of a low-ranked product, and turning regular customers away can push them to competitors for good. The simple ranking also assumes one constraint; when two or more resources are scarce at once, firms use linear programming to find the best mix. Demand forecasts are estimates, so plans should be revisited as orders arrive.
Common mistakes to avoid
1. Ranking products by contribution margin per unit, or by selling price, instead of by margin per unit of the constraint. 2. Including fixed costs, which do not change with the product mix. 3. Giving a product more hours than its demand can use. 4. Valuing an extra hour at the top-ranked product's rate after its demand is already met. 5. Adding capacity to a resource that is not the bottleneck. 6. Ignoring product sets, contracts and customer relationships.
Key takeaways
When a scarce resource limits output, fixed costs are irrelevant and the aim is the largest total contribution margin. Rank products by contribution margin per unit of the constrained resource, then fill demand in that order until the resource runs out. An extra unit of the constraint is worth the margin per unit of the constraint of the best product with unmet demand, which caps what the firm should pay for overtime or extra capacity. Relax the bottleneck itself, and check the plan against product sets, contracts and customer needs.
Flashcards
Flashcard
A scarce resource that limits total output
Test your understanding
Question 1 of 5
Machine hours are scarce and demand is unlimited. Product X earns a contribution margin of $30 and needs 3 machine hours; product Y earns $20 and needs 1 hour. Which should be made first?
Question 2 of 5
A firm has 1,000 machine hours. Product P earns $40 a unit, needs 2 hours, and has demand of 300 units. Product Q earns $25 a unit, needs 1 hour, and has demand of 500 units. What is the largest total contribution margin it can earn?
Question 3 of 5
In the plan that maximises margin for the firm above, product P still has unmet demand. What is the most worth paying in extra costs for one more machine hour?
Question 4 of 5
Which action relaxes a production constraint?
Question 5 of 5
Why might a firm still make some of its lowest-ranked product when a resource is scarce?
