CHAPTER 20 INVENTORY MANAGEMENT, JUST-IN-TIME, AND...

30
20-1 CHAPTER 20 INVENTORY MANAGEMENT, JUST-IN-TIME, AND SIMPLIFIED COSTING METHODS 20-1 Cost of goods sold (in retail organizations) or direct materials costs (in organizations with a manufacturing function) as a percentage of sales frequently exceeds net income as a percentage of sales by many orders of magnitude. In the Kroger grocery store example cited in the text, cost of goods sold to sales is 73.7%, and net income to sales is 0.6%. Thus, a 10% reduction in the ratio of cost of goods sold to sales (73.7 to 66.3%) without any other changes can result in a 1233% increase in net income to sales (0.6% to 8.0%). 20-2 Five cost categories important in managing goods for sale in a retail organization are the following: 1. purchasing costs; 2. ordering costs; 3. carrying costs; 4. stockout costs; and 5. quality costs 20-3 Five assumptions made when using the simplest version of the EOQ model are: 1. The same quantity is ordered at each reorder point. 2. Demand, ordering costs, carrying costs, and the purchase-order lead time are certain. 3. Purchasing cost per unit is unaffected by the quantity ordered. 4. No stockouts occur. 5. Costs of quality are considered only to the extent that these costs affect ordering costs or carrying costs. 20-4 Costs included in the carrying costs of inventory are incremental costs for such items as insurance, rent, obsolescence, spoilage, and breakage plus the opportunity cost of capital (or required return on investment). 20-5 Examples of opportunity costs relevant to the EOQ decision model but typically not recorded in accounting systems are the following: 1. the return forgone by investing capital in inventory; 2. lost contribution margin on existing sales when a stockout occurs; and 3. lost contribution margin on potential future sales that will not be made to disgruntled customers. 20-6 The steps in computing the costs of a prediction error when using the EOQ decision model are: Step 1: Compute the monetary outcome from the best action that could be taken, given the actual amount of the cost input. Step 2: Compute the monetary outcome from the best action based on the incorrect amount of the predicted cost input. Step 3: Compute the difference between the monetary outcomes from Steps 1 and 2.

Transcript of CHAPTER 20 INVENTORY MANAGEMENT, JUST-IN-TIME, AND...

20-1

CHAPTER 20 INVENTORY MANAGEMENT, JUST-IN-TIME,

AND SIMPLIFIED COSTING METHODS 20-1 Cost of goods sold (in retail organizations) or direct materials costs (in organizations with a manufacturing function) as a percentage of sales frequently exceeds net income as a percentage of sales by many orders of magnitude. In the Kroger grocery store example cited in the text, cost of goods sold to sales is 73.7%, and net income to sales is 0.6%. Thus, a 10% reduction in the ratio of cost of goods sold to sales (73.7 to 66.3%) without any other changes can result in a 1233% increase in net income to sales (0.6% to 8.0%). 20-2 Five cost categories important in managing goods for sale in a retail organization are the following: 1. purchasing costs; 2. ordering costs; 3. carrying costs; 4. stockout costs; and 5. quality costs 20-3 Five assumptions made when using the simplest version of the EOQ model are: 1. The same quantity is ordered at each reorder point. 2. Demand, ordering costs, carrying costs, and the purchase-order lead time are certain. 3. Purchasing cost per unit is unaffected by the quantity ordered. 4. No stockouts occur. 5. Costs of quality are considered only to the extent that these costs affect ordering costs or

carrying costs. 20-4 Costs included in the carrying costs of inventory are incremental costs for such items as insurance, rent, obsolescence, spoilage, and breakage plus the opportunity cost of capital (or required return on investment). 20-5 Examples of opportunity costs relevant to the EOQ decision model but typically not recorded in accounting systems are the following: 1. the return forgone by investing capital in inventory; 2. lost contribution margin on existing sales when a stockout occurs; and 3. lost contribution margin on potential future sales that will not be made to disgruntled

customers. 20-6 The steps in computing the costs of a prediction error when using the EOQ decision model are:

Step 1: Compute the monetary outcome from the best action that could be taken, given the actual amount of the cost input.

Step 2: Compute the monetary outcome from the best action based on the incorrect amount of the predicted cost input.

Step 3: Compute the difference between the monetary outcomes from Steps 1 and 2.

20-2

20-7 Disagree. The optimal safety stock level is the quantity of safety stock that minimizes the sum of the annual relevant stockout costs and carrying costs. 20-8 Just-in-time (JIT) purchasing is the purchase of materials (or goods) so that they are delivered just as needed for production (or sales). Benefits include lower inventory holdings (reduced warehouse space required and less money tied up in inventory) and less risk of inventory obsolescence and spoilage. 20-9 Factors causing reductions in the cost to place purchase orders of materials are:

• Companies are establishing long-run purchasing agreements that define price and quality terms over an extended period.

• Companies are using electronic links, such as the Internet, to place purchase orders. • Companies are increasing the use of purchase-order cards.

20-10 Disagree. Choosing the supplier who offers the lowest price will not necessarily result in the lowest total purchase cost to the buyer. This is because the price or purchase cost of the goods is only one—and perhaps, most obvious—element of cost associated with purchasing and managing inventories. Other relevant cost items are ordering costs, carrying costs, stockout costs and quality costs. A low-cost supplier may well impose conditions on the buyer—such as poor quality, or frequent stockouts, or excessively high inventories—that result in high total costs of purchase. Buyers must examine all the elements of costs relevant to inventory management, not just the purchase price. 20-11 The supply chain describes the flow of goods, services, and information from the initial sources of materials and services to the delivery of products to consumers, regardless of whether those activities occur in the same organization or in other organizations. Sharing of information across companies enables a reduction in inventory levels at all stages, fewer stockouts at the retail level, reduced manufacture of product not subsequently demanded by retailers, and a reduction in expedited manufacturing orders. 20-12 Just-in-time (JIT) production is a “demand-pull” manufacturing system that has the following features:

• Organize production in manufacturing cells, • Hire and retain workers who are multi-skilled, • Aggressively pursue total quality management (TQM) to eliminate defects, • Place emphasis on reducing both setup time and manufacturing lead time, and • Carefully select suppliers who are capable of delivering quality materials in a timely

manner. 20-13 The main reason why backflush costing does not strictly adhere to the GAAP is that the work-in-process accounts are not recognized in the accounting records. Work in process consists of unfinished goods. Substantial business resources were dedicated to their production, and should be recognized in the accounts as an asset.

20-3

20-14 Versions of backflush costing differ in the number and placement of trigger points at which journal entries are made in the accounting system:

Number of Journal Entry Trigger Points

Location in Cycle Where

Journal Entries Made Version 1 3 Stage A. Purchase of direct materials Stage C. Completion of good finished units of product Stage D. Sale of finished goods Version 2 2 Stage A. Purchase of direct materials Stage D. Sale of finished goods Version 3 2 Stage C. Completion of good finished units of product Stage D. Sale of finished goods

20-15 Traditional accounting systems cost individual products, and separate product costs from selling, general, and administrative costs. Lean accounting costs the entire value stream instead of individual products. Unused capacity costs and common costs that cannot be reasonably assigned to value streams are excluded from value stream costs. In addition, many lean accounting users expense material costs the period they are purchased, rather than storing them on the balance sheet until the products using the material are sold.

20-4

20-16 (20 min.) Economic order quantity for retailer. 1. D = 10,000, P = $200, C = $7

7

200$000,102

C

DP 2 EOQ

××==

= 755.93 ≅ 756 jerseys

2. Number of orders per year = EOQ

D =

756

000,10

= 13.22 ≅ 14 orders

3. day workingeach Demand =

days workingofNumber

D

= 365

000,10

= 27.40 jerseys per day Purchase lead time = 7 days Reorder point = 27.40 × 7 = 191.80 ≅ 192 jerseys

20-5

20-17 (20 min.) Economic order quantity, effect of parameter changes (continuation of 20-16).

1. D = 10,000, P = $25, C = $7

7

25$000,102

C

DP 2 EOQ

××==

= 267.26 jerseys ≅ 267 jerseys

The sizable reduction in ordering cost (from $200 to $25 per purchase order) has reduced the EOQ from 756 to 267.

2. The AT proposal has both upsides and downsides. The upside is potentially higher sales. FB customers may purchase more online than if they have to physically visit a store. FB would also have lower administrative costs and lower inventory holding costs with the proposal.

The downside is that AT could capture FB’s customers. Repeat customers to the AT web site need not be classified as FB customers. FB would have to establish enforceable rules to make sure it captures ongoing revenues from customers it directs to the AT web site.

There is insufficient information to determine whether FB should accept AT’s proposal. Much depends on whether FB views AT as a credible, “honest” partner.

20-18 (15 min.) EOQ for a retailer. 1. D = 20,000, P = $160, C = 20% × $8 = $1.60

EOQ = 2DP

C = 2 × 20‚000 × $160

$1.60 = 2,000 yards

2. Number of orders per year: D

EOQ = 20‚0002‚000 = 10 orders

3. Demand each working day = D

Number of working days

= 20‚000

250

= 80 yards per day = 400 yards per week Purchasing lead time = 2 weeks Reorder point = 400 yards per week × 2 weeks = 800 yards

20-6

20-19 (20 min.) EOQ for manufacturer. 1. Relevant carrying costs per part per year: Required annual return on investment 20% × $50 = $ 10

Relevant insurance, materials handling, breakage, etc. costs per year 5 Relevant carrying costs per part per year $15 With D = 15,000; P = $125; C = $15, EOQ for manufacturer is:

2DPC =

2 15,000 $125

$15

× ×= 500 units

2. Total relevant ordering costs = ⎟⎟

⎞⎜⎜⎝

⎛× P

Q

D

= ⎟⎠⎞

⎜⎝⎛ ×$125

500

000,15

= $3,750 where Q = 500 units, the EOQ. 3. At the EOQ, total relevant ordering costs and total relevant carrying costs will be exactly

equal. Therefore, total relevant carrying costs at the EOQ = $3,750 (from requirement 2). We can also confirm this with direct calculation:

costs carryingrelevant Total = ⎟⎠⎞

⎜⎝⎛ ×C

2

Q

= ⎟⎠

⎞⎜⎝

⎛ ×$15 2

500

= $3,750 where Q = 500 units, the EOQ. 4. Purchase order lead time is half a month. Monthly demand is 15,000 units ÷ 12 months = 1,250 units per month.

Demand in half a month is 12 × 1,250 units or 625 units.

Greenview should reorder when the inventory of rotor blades falls to 625 units.

20-7

20-20 (20 min.) Sensitivity of EOQ to changes in relevant ordering and carrying costs. 1. A straightforward approach to the requirement is to construct the following table for

EOQ at relevant carrying and ordering costs. Annual demand is 10,000 units. The formula for the EOQ model is:

EOQ = 2DP

C

where D = demand in units for a specified period of time

P = relevant ordering costs per purchase order

C = relevant carrying costs of one unit in stock for the time period used for D (one year in this problem.

Relevant Carrying Costs per Unit Relevant Ordering Costs per Purchase Order

per Year $300 $200

$10 10$

300$000,102 ××= 775

10$

200$000,102 ××= 632

15 15$

300$000,102 ××= 632

15$

200$000,102 ××= 516

20 20$

300$000,102 ××= 548

20$

200$000,102 ××= 447

2. For a given demand level, as relevant carrying costs increase, EOQ becomes smaller. For

a given demand level, as relevant order costs increase, EOQ increases.

20-8

20-21 (15 min.) Inventory management and the balanced scorecard. 1. The incremental increase in operating profits from employee cross-training (ignoring the

cost of the training) is: Increased revenue from higher customer satisfaction ($6,000,000 x 2% x 5) $600,000 Reduced inventory-related costs 120,000 Incremental increase in operating profits (ignoring training costs) $720,000 2. At a cost of $720,000, SSC will be indifferent between current expenditures and

increasing employee cross-training by 5%. Consequently, the most SSC would be willing to pay for this cross-training is the $720,000 benefit received.

3. Besides increasing short-term operating profits, additional employee cross-training can improve employee satisfaction because their jobs can have more variety, potentially leading to unanticipated productivity improvements and lower employee turnover. Multi-skilled employees can also understand the production process better and can suggest potential improvements. Each of these may lead to additional cost reductions.

20-9

20-22 (20 min.) JIT production, relevant benefits, relevant costs. 1. Solution Exhibit 20-22 presents the annual net benefit of $315,000 to Champion

Hardware Company of implementing a JIT production system. 2. Other nonfinancial and qualitative factors that Champion should consider in deciding

whether it should implement a JIT system include: a. The possibility of developing and implementing a detailed system for integrating the

sequential operations of the manufacturing process. Direct materials must arrive when needed for each subassembly so that the production process functions smoothly.

b. The ability to design products that use standardized parts and reduce manufacturing time.

c. The ease of obtaining reliable vendors who can deliver quality direct materials on time with minimum lead time.

d. Willingness of suppliers to deliver smaller and more frequent orders. e. The confidence of being able to deliver quality products on time. Failure to do so

would result in customer dissatisfaction. f. The skill levels of workers to perform multiple tasks such as minor repairs,

maintenance, quality testing and inspection. SOLUTION EXHIBIT 20-22 Annual Relevant Costs of Current Production System and JIT Production System for Champion Hardware Company

Relevant Items

Relevant

Costs under Current

Production System

Relevant

Costs under JIT

Production System

Annual tooling costs – $100,000 Required return on investment: 15% per year × $1,000,000 of average inventory per year $150,000 15% per year × $200,000a of average inventory per year 30,000 Insurance, space, materials handling, and setup costs 300,000 225,000b Rework costs 200,000 140,000c Incremental revenues from higher selling prices (160,000)d Total net incremental costs $650,000 $335,000 Annual difference in favor of JIT production $315,000 a $1,000,000 × (1 – 80%) = $200,000 b$300,000 × (1 – 0.25) = $225,000

c$200,000 × (1 – 0.30) = $140,000

d$4 × 40,000 units = $160,000

20-10

3. Personal observation by production line workers and managers is more effective in JIT plants than in traditional plants. A JIT plant’s production process layout is streamlined. Operations are not obscured by piles of inventory or rework. As a result, such plants are easier to evaluate by personal observation than cluttered plants where the flow of production is not logically laid out.

Besides personal observation, nonfinancial performance measures are the dominant methods of control. Nonfinancial performance measures provide most timely and easy to understand measures of plant performance. Examples of nonfinancial performance measures of time, inventory, and quality include: • Manufacturing lead time • Units produced per hour • Machine setup time ÷ manufacturing time • Number of defective units ÷ number of units completed

In addition to personal observation and nonfinancial performance measures, financial performance measures are also used. Examples of financial performance measures include: • Cost of rework • Ordering costs • Stockout costs • Inventory turnover (cost of goods sold ÷ average inventory)

The success of a JIT system depends on the speed of information flows from customers to manufacturers to suppliers. The Enterprise Resource Planning (ERP) system has a single database, and gives lower-level managers, workers, customers, and suppliers access to operating information. This benefit, accompanied by tight coordination across business functions, enables the ERP system to rapidly transmit information in response to changes in supply and demand so that manufacturing and distribution plans may be revised accordingly.

20-11

20-23 (30 min.) Backflush costing and JIT production.

1. (a) Purchases of direct materials Inventory: Materials and In-Process

Control

2,754,000

Accounts Payable Control 2,754,000 (b) Incur conversion costs Conversion Costs Control 723,600 Various Accounts 723,600 (c) Completion of finished goods Finished Goods Controla 3,484,000 Inventory: Materials and In-Process

Control

2,733,600 Conversion Costs Allocated 750,400 (d) Sale of finished goods Cost of Goods Soldb 3,432,000 Finished Goods Control 3,432,000 a26,800 × ($102 + $28) = $3,484,000 b26,400 × ($102 + $28) = $3,432,000

20-12

2.

Inventory:

3.

Under an ideal JIT

production system, there could be zero inventories at the end of each day. E

ntry (c) would be $3,432,000

finished goods production, not $3,484,000. Also, there w

ould be no inventory of direct materials instead of $2,754,000 –

$2,733,600 = $20,400.

Finished Goods C

ontrol C

ost of Goods Sold

Materials and In-P

rocess Control

(b) 723,600

Conversion C

osts Control

(c) 750,400

Conversion C

osts Allocated

(d) 3,432,000 (d

) 3,432,000 (c) 3,484,000

Bal. 20,400

(a) 2,754,000 (c) 2,733,600

Bal. 52,000

Conversion

Costs

Direct

Materials

20-13

20-24 (20 min.) Backflush costing, two trigger points, materials purchase and sale (continuation of 20-23).

1. (a) Purchases of direct materials Inventory Control 2,754,000 Accounts Payable Control 2,754,000 (b) Incur conversion costs Conversion Costs Control 723,600 Various Accounts 723,600 (c) Completion of finished goods No entry (d) Sale of finished goods Cost of Goods Sold 3,432,000 Inventory Control 2,692,800 Conversion Costs Allocated 739,200 (e) Underallocated or Conversion Costs Allocated 739,200 overallocated conversion Costs of Goods Sold 15,600 costs Conversion Costs Control 723,600 2.

(e) 15,600 (d) 3,432,000

Cost of Goods Sold Inventory Control

(d) 2,692,800 (a) 2,754,000

Bal. 61,200

(d) 739,200

Conversion Costs Allocated

(e) 739,200

Conversion Costs Control

(b) 723,600 (e) 723,600

Conversion Costs

Direct Materials

20-14

20-25 (20 min.) Backflush costing, two trigger points, completion of production and sale (continuation of 20-23).

(a) Purchases of direct materials No Entry (b) Incur conversion costs Conversion Costs Control 723,600 Various Accounts 723,600 (c) Completion of finished goods Finished Goods Control 3,484,000 Accounts Payable Control 2,733,600 Conversion Costs Allocated 750,400 (d) Sale of finished goods Cost of Goods Sold 3,432,000 Finished Goods Control 3,432,000 (e) Underallocated or Conversion Costs Allocated 750,400 Overallocated conversion Costs of Goods Sold 26,800 Costs Conversion Costs Control 723,600

{

(c) 750,400

Conversion Costs Allocated

(e) 750,400

(b) 723,600

Conversion Costs Control

(e) 723,600

(d) 3,432,000 (e) 26,800

Cost of Goods Sold

(c) 3,484,000

Finished Goods Control

(d) 3,432,000

Bal. 52,000

Conversion Costs

Direct Materials

20-15

20-26 (30 m

in.) E

ffect of different order quantities on ordering costs and carrying costs, EO

Q.

1.

Scenario

1

2 3

4 5

Dem

and (units) (D)

234,000234,000

234,000 234,000

234,000C

ost per purchase order (P)

$ 81.00$ 81.00

$ 81.00 $ 81.00

$ 81.00A

nnual carrying cost per package (C)

$ 11.70$ 11.70

$ 11.70 $ 11.70

$ 11.70O

rder quantity per purchase order (units) (Q)

9001,500

1,800 2,100

2,700N

umber of purchase orders per year (D

÷ Q

) 260.00

156.00130.00

111.4386.67

Annual ordering costs (D

÷Q

) ×P

$21,060$12,636

$10,530 $ 9,026

$ 7,020A

nnual carrying costs (QC

÷ 2)

$ 5,265$ 8,775

$10,530 $12,285

$15,795T

otal relevant costs of ordering and carrying inventory $26,325

$21,411 $21,060

$21,311 $22,815

T

he economic order quantity is 1,800 packages. It is the order quantity at w

hich carrying costs equal ordering costs and total relevant ordering and carrying costs are m

inimized.

W

e can also confirm this from

direct calculation. Using D

= 234,000; P =

$81 and C =

$11.70

E

OQ

=

70.

11$

81$

000,

2342

××

= 1,800 packages

It is interesting to note that Koala B

lue faces a situation where total relevant ordering and carrying costs do not vary very m

uch w

hen order quantity ranges from 1,500 packages to 2,700 packages.

20-16

2. When the ordering cost per purchase order is reduced to $49:

EOQ = 70.11$

49$000,2342 ×× = 1,400 packages

The EOQ drops from 1,800 packages to 1,400 packages when Koala Blue’s ordering cost per purchase order decreases from $81 to $49.

And the new relevant costs of ordering inventory = ⎟⎟⎠

⎞⎜⎜⎝

⎛× P

Q

D = ⎟

⎞⎜⎝

⎛×$49

400,1

000,234 = $8,190

and the new relevant costs or carrying inventory = ⎟⎠⎞

⎜⎝⎛ ×C

2

Q = ⎟

⎠⎞

⎜⎝⎛ ×$11.70

2

400,1 = $8,190

The total new costs of ordering and carrying inventory = $8,190 × 2 = $16,380 3. As summarized below, the new Mona Lisa web-based ordering system, by lowering the

EOQ to 1,400 packages, will lower the carrying and ordering costs by $4,680. Koala Blue will spend $2,000 to train its purchasing assistants on the new system. Overall, Koala Blue will still save $2,680 in the first year alone.

Total relevant costs at EOQ (from Requirement 2) $16,380 Annual cost benefit over old system ($21,060 – $16,380) $ 4,680 Training costs 2,000 Net benefit in first year alone $ 2,680

20-17

20-27 (30 min.) EOQ, uncertainty, safety stock, reorder point.

1. 8.4

500$000,2402

C

DP 2 EOQ

××==

= 7,071 pairs of shoes 2. If monthly demand averages 20,000 pairs of shoes (240,000/12 months) then

Warehouse MC3 would place orders three times a month (20,000/7,071 = 2.83 = 3 times) 3. Weekly demand (reorder point) = Monthly Demand / 4 = 20,000/4 = 5,000 pairs of shoes 4. Safety Stock = 20% x 20,000 = 4,000 pairs of shoes

Reorder Point = Weekly Demand + Safety Stock = 5,000 + 4,000 = 9,000 pairs of shoes

EOQ = 7,071 pairs of shoes (since neither annual demand, ordering cost, nor carrying

cost have changed, the EOQ will not change). 5.

Without Safety Stock With Safety Stock

⎟⎠

⎞⎜⎝

⎛×= P

Q

Dcostsordering

relevantTotal

⎟⎟⎠

⎞⎜⎜⎝

⎛×$500

071,7

000,240= $16,971 ⎟⎟

⎞⎜⎜⎝

⎛×$500

071,7

000,240= $16,971

costs carrying relevant Total =

⎟⎠⎞

⎜⎝⎛ ×+ C Stock)Safety

2

Q(

⎟⎠

⎞⎜⎝

⎛ ×$4.80 2

071,7= $16,970 ⎟

⎠⎞

⎜⎝⎛ ×+ $4.80 0)00,4

2

071,7(

= $36,170

Total Relevant Cost

$33,941

$53,141

20-18

20-28 (25 min.) MRP, EOQ, and JIT. 1. Under a MRP system:

Annual cost of producing and carrying J-Pods in inventory = Variable Production Cost + Setup Cost + Carrying Cost = $50 x 48,000 + ($50,000 x 12 months) + [$20 x (4,000/2)]

= $2,400,000 + 600,000 + 40,000 = $3,040,000

2. Using an EOQ model to determine batch size:

20

000,50$000,482

C

DP 2 EOQ

××==

= 15,492 J-Pods per batch

Production of 48,000 per year divided by a batch size of 15,492 would imply J-Pods would be produce 3.1 batches per year. Rounding this up to the nearest whole number yields 4 batches per year.

Annual Cost of producing and carrying J-Pods in inventory = Variable Production Cost + Setup Cost + Carrying Cost = $50 x 48,000 + ($50,000 x 4) + [$20 x (15,492/2)]

= $2,400,000 + 200,000 + 154,920 = $2,754,920

3. Under a JIT system

Annual Cost of producing and carrying J-Pods in inventory = Variable Production Cost + Setup Cost + Carrying Cost = $50 x 48,000 + ($5,000 x 96 a) + [$20 x (500/2)]

= $2,400,000 + 480,000 + 5,000 = $2,885,000 a production of 48,000 per year divided by a batch size of 500 would imply materials would be ordered 96 times per year.)

4. The EOQ system resulted in the lowest costs, despite the fact that carrying costs were

lower for the JIT model. However, the EOQ model, in this case, limits production to only once every four months. This would not allow managers to react quickly to changing market demand or economic conditions. The JIT model provides management with much more flexibility.

20-19

20-29 (30 min.) Effect of management evaluation criteria on EOQ model. 1.

50

800$000,5002

C

DP 2 EOQ

××==

= 4,000 computers

2. Number of orders per year = EOQ

D =

000,4000,500

= 125 orders

Total relevant ordering costs = ⎟

⎞⎜⎝

⎛× P

QD

= ⎟⎠

⎞⎜⎝

⎛ ×$800 000,4000,500

= $100,000

costs carryingrelevant Total = ⎟⎠⎞

⎜⎝⎛ ×C

2Q

= ⎟⎠⎞

⎜⎝⎛ ×$50

2000,4

= $100,000 Total Relevant Cost = $100,000 + 100,000 = $200,000

3.

30

800$000,5002

C

DP 2 EOQ

××==

= 5,164 computers

Total relevant ordering costs = ⎟

⎞⎜⎝

⎛× P

QD

= ⎟⎠

⎞⎜⎝

⎛ ×$800 164,5

000,500

= $77,459

20-20

costs carryingrelevant Total = ⎟⎠⎞

⎜⎝⎛ ×C

2Q

= ⎟⎠⎞

⎜⎝⎛ ×$50

2164,5

= $129,100

Total relevant cost = $77,459 + 129,100 = $206,559

Thus, the EOQ quantity and total relevant costs are higher if the company ignores holding costs when evaluating managers.

4. Since managers will choose to order 5,164 computers instead of 4,000, the cost to the

company will be $6, 559 ($206,559 - $200,000) higher than it would be if managers were evaluated based upon all carrying costs. Computers 4 U probably does not include the opportunity costs of carrying inventory because it is not tracked by the financial accounting system. The company could change the evaluation model to include a return on investment on inventory. Even though this would involve an additional calculation, it would encourage managers to make optimal decisions.

20-21

20-30 (25 min.) Effect of EOQ ordering on supplier costs (continuation of Problem 20-29). 1.

i) Set up cost = Cost per setup x annual setups

Alternative A: $1,000 x 50 setups = $50,000 Alternative B: $1,000 x 250 setups = $250,000

ii) Carrying Cost = Average inventory level x carrying cost

Alternative A: 10,000/2 x $50 = $250,000 Alternative B: Assumed to be $0

iii) Total relevant cost

Alternative A: $50,000 + $250,000 = $300,000 Alternative B: $250,000 + $0 = $250,000

Costs would be lower if IMBest produced computers every day. 2. Total cost, where C = carrying costs per unit.

Alternative A: $50,000 + (10,000/2) x C Alternative B: $250,000 + $0 Indifference point: $50,000 + $5,000C = $250,000

$5,000C = $200,000 C = $40

If carrying costs fall below $40 per unit, IMBest would be better off producing the computers once a week.

20-22

20-31 (30 min.) JIT purchasing, relevant benefits, relevant costs. 1. Solution Exhibit 20-31 presents the $50,500 cash savings that would result if Trenton

Corporation adopted the just-in-time inventory system in 2008. 2. Conditions that should exist in order for a company to successfully adopt just-in-time

purchasing include the following:

• Top management must be committed and provide the necessary leadership support to ensure a company-wide, coordinated effort.

• A detailed system for integrating the sequential operations of the manufacturing process needs to be developed and implemented. Direct materials must arrive when needed for each subassembly so that the production process functions smoothly.

• Accurate sales forecasts must be available for effective finished goods planning and production scheduling.

• Products should be designed to maximize use of standardized parts to reduce manufacturing time and costs.

• Reliable vendors who can deliver quality direct materials on time with minimum lead time must be obtained.

20-23

SOLUTION EXHIBIT 20-31 Annual Relevant Costs of Current Purchasing Policy and JIT Purchasing Policy for Trenton Corporation

Relevant Costs under

Current Purchasing

Policy

Relevant Costs under

JIT Purchasing

Policy Required return on investment 20% per year × $800,000 of average inventory per year $160,000 20% per year × $0 inventory per year $ 0 Annual insurance and property tax costs 15,000 0 Warehouse rent 70,000 (18,000)a Overtime costs No overtime 0 Overtime premium 50,000 Stockout costs No stockouts 0 $6.50b contribution margin per unit ×25,000 units 162,500 Total incremental costs $245,000 $194,500 Difference in favor of JIT purchasing $50,500 a$(18,000) = Warehouse rental revenues, [(75% × 15,000) × $1.60]. bCalculation of unit contribution margin

Selling price ($12,960,000 ÷ 1,080,000 units) $12.00

Variable costs per unit : Variable manufacturing cost per unit

($4,860,000 ÷ 1,080,000 units) $4.50 Variable marketing and distribution cost per unit

($1,080,000 ÷ 1,080,000000 units) 1.00 Total variable costs per unit 5.50 Contribution margin per unit $ 6.50

Note that the incremental costs of $50,000 in overtime premiums to make the additional 15,000 units are less than the contribution margin from losing these sales equal to $97,500 ($6.50 × 15,000). Trenton would rather incur overtime than lose 15,000 units of sales.

20-24

20-32 (25 min.) Supply chain effects on total relevant inventory costs.

1. The relevant costs of purchasing from Maji and Induk.

Cost Category Maji Induk Purchase Cost 10,000 x $93 10,000 x $90

$930,000

$900,000

Order Cost 50 x $10 50 x $8

$500

$400

Inspection Cost 10,000 x $5 x 5% 10,000 x $5 x 25%

$2,500

$12,500

Required ROI 100 units x $93 x 10% 100 units x $90 x 10%

$930

$900

Stockout Cost 100 x $5 300 x $8

$500

$2,400

Return Cost 50 x $25 500 x $25

$1,250

$12,500

Other Carrying Cost 100 x $2.50 100 x $2.50

$250

$250

Total Cost $935,930 $928,950

2. While Induk will save Cow Spot $6,980, Cow Spot may still choose to use Maji for the following reasons:

a. The savings are less than 1% of the total cost of the mother boards. b. With ten times the number of returns, Induk will probably have a negative effect

on Cow Spot’s reputation. c. With Induk’s higher stockouts, Cow Spot’s reputation for availability and on time

delivery will be effected. d. The increased number of inspections may necessitate the hiring of additional

personnel and the need for additional factory space and equipment.

20-25

20-33 (20 min.) Blackflush costing and JIT production. 1. (a) Purchases of direct materials Inventory: Materials and In-Process Control 550,000 Accounts Payable Control 550,000 (b) Incur conversion costs Conversion Costs Control 440,000 Various Accounts 440,000 (c) Completion of finished goods Finished Goods Controla 945,000 Inventory: Materials & In-Process Control 525,000 Conversion Costs Allocated 420,000 (d) Sale of finished goods Cost of Goods Soldb 900,000 Finished Goods Control 900,000 a

21,000 × $45 ($25 + $20) = $945,000 b

20,000 × $45 = $900,000

2.

( b ) 4 4 0 , 0 0 0

( a ) 5 5 0 , 0 0 0 ( c ) 5 25,000 (c) 945,000

B al. 2 5 , 0 0 0

( c ) 4 20,000

(d) 900,000

Bal. 45,000

( d ) 9 0 0 , 0 0 0

Inventory Materials and In-Process Control Finished Goods Control Cost of Goods Sold

Conversion Costs Allocated

Conversion Costs Control

Conversion Costs

Direct Materials

20-26

20-34 (20 min.) Backflush, two trigger points, materials purchase and sale (continuation of 20-33).

1. (a) Purchases of direct materials Inventory Control 550,000 Accounts Payable Control 550,000 (b) Incur conversion costs Conversion Costs Control 440,000 Various Accounts (such as Accounts Payable Control and Wages Payable Control) 440,000 (c) Completion of finished goods No entry (d) Sale of finished goods Cost of Goods Sold 900,000 Inventory Control 500,000 Conversion Costs Allocated 400,000 (e) Underallocated Conversion Costs Allocated 400,000 or overallocated Cost of Goods Sold 40,000 conversion costs Conversion Costs Control 440,000 2.

(d) 900,000

(b) 440,000 (e) 440,000

Conversion Costs

Inventory Control

Conversion Costs Allocated

Conversion Costs Control

Cost of Goods Sold

(a) 550,000 (d) 500,000

(e) 400,000 (d) 400,000

Bal.50,000

(e) 40,000

Direct Materials

20-27

20-35 (20 min.) Backflush, two trigger points, completion of production and sale (continuation of 20-33). 1. (a) Purchase of direct materials No entry (b) Incur conversion costs Conversion Costs Control 440,000 Various Accounts (such as Accounts Payable Control and Wages Payable Control) 440,000 (c) Completion of finished Finished Goods Control 945,000 goods Accounts Payable Control 525,000 Conversion Costs Allocated 420,000 (d) Sale of finished goods Cost of Goods Sold 900,000 Finished Goods Control 900,000 (e) Underallocated Conversion Costs Allocated 420,000 or overallocated Cost of Goods Sold 20,000 conversion costs Conversion Costs Control 440,000 2.

D i r e c t M a t e r i a l s

C o n v e r s i o n C o s t s

C o nve r s i o n C o s t s A l l ocated( c ) 4 2 0,000

C onv e r s i o n C o s t s C ontrol

( b) 4 4 0 , 0 0 0

( e) 4 2 0 , 0 0 0

( e ) 4 4 0,000

(c) 945,000 (d) 900,000 ( d ) 9 0 0 , 0 0 0

( e ) 2 0 , 0 0 0

Finished Goods Control Cost of Goods Sold

Bal. 45,000

20-28

20-36 (20 min.) Lean accounting.

The cost object in lean accounting is the value stream, not the individual product. FSD has identified two distinct value streams: Mechanical Devices and Electronic Devices. All direct costs are traced to the value streams. However, not all plant-level overhead costs are allocated to the value streams when computing operating income. Since value streams are only charged for the percentage of space they actually use, only 85% of the $180,000 occupancy costs are charged to the two value streams. The remaining 15%, or $18,000, is not used to compute value stream profits.

2. Operating income under lean accounting are the following (in thousands of dollars):

Mechanical Devices

Electronic Devices

Sales $1,200 $1,350 - Costs Direct material purchased 315 340 Direct labor 225 260 Equipment costs 215 300 Product-line overhead 170 175 Occupancy costs ($120,000 x 40%) ($120,000 x 45%)

48

54

Value stream operating income $227 $221

In addition to the differences discussed in Requirement 1, FSD’s lean accounting system treats all direct material expenditures as expenses in the period they are purchased. The following factors explain the differences between traditional operating income and lean accounting income for the two value streams:

Mechanical

Devices Electronic

Devices Traditional operating income $205 $185 Additional cost of direct materials (15) (15) Decrease in allocated plant-level overhead 37 51

Value stream operating income $227 $221

20-29

20-37 (35–40 min.) Backflushing.

1. Greenwood has successfully implemented JIT in its production operations and, as a result, minimized work-in-process inventory. However, it still has a fair amount of raw material and finished goods inventory. Greenwood should, therefore, adopt a backflush costing system with two trigger points, as follows: a. Direct materials purchases charged to Inventory: Materials and In-Process Control b. Completion of finished goods recorded as Finished Goods Control

The backflush approach described closely approximates the costs computed using sequential tracking. There is no work in process so there is no need for a Work in Process inventory account.

Further, by maintaining a Materials and In-Process Inventory Control and Finished Goods Control account, Greenwood can keep track of and control the inventories of direct materials and finished goods in its plant.

2a. Greenwood should adopt a backflush costing system with trigger points at completion of finished goods and at the sale of finished goods. This would approximate the sequential tracking approach since the question assumes Greenwood has no direct materials or work-in-process inventories. There is, therefore, no need for these inventory accounts.

b. A backflush costing system with two trigger points—when purchases of direct materials are made (debited to Inventory Control), and when Finished Goods are sold—would approximate sequential tracking, since the question assumes Greenwood has no work-in-process or finished goods inventories.

c. A backflush costing system with a single trigger point when finished goods are sold would approximate sequential tracking, since the question assumes Greenwood has no direct material, work-in-process or finished goods inventories. This is a further simplification of the examples in the text.

The principle here is that backflushing of costs should be triggered at the finished goods inventory stage if Greenwood plans to hold finished goods inventory. If Greenwood plans to hold no finished goods inventory, backflushing can be postponed until the finished goods are sold. In other words, the trigger points for backflushing relate to the points where inventory is being accumulated. As a result, backflushing matches the sequential tracking approach and also maintains a record for the monitoring and control of the inventory.

3. Some comments on the quotation follow: a. The backflush system is a standard costing system, not an actual costing system. b. If standard costing is used, an up-to-date, realistic set of standard costs is always

desirable––as long as the set meets the cost-benefit test of updating. c. The operating environments of “the present JIT era” have induced many companies

toward more simplicity (backflush) and abandoning the typical standard costing system (sequential tracking).

d. Backflush is probably closer to being a periodic system than a perpetual system. However, a periodic system may be cost-effective, particularly where physical inventories are relatively low or stable.

20-30

e. The textbook points out that, to be attractive, backflush costing should generate the same financial measurements as sequential tracking––and at a lower accounting cost.

f. The choice of a product costing system is highly contextual. Its characteristics should be heavily affected by its costs, the preferences of operating managers, and the underlying operating processes. Sweeping generalizations about any cost accounting system or technique are unjustified.