9

C H A P T E R

Layout Strategies

DISCUSSION QUESTIONS  Individual workloads can usually be made to be

 1. The seven layout strategies are: relatively equivalent.
 Fixed position or project layout  It usually results in low variable cost per unit.
 Process-oriented layout  It tends to keep material handling costs relatively low.
 Warehouse layout  It reduces work-in-process inventories.
 Retail layout
 It makes training and supervision easier than with
 Product-oriented/repetitive layout other layout strategies.
 Office layout (b) The disadvantages are:
 Work cells  High volume is required because of the large
 2. Fixed position layouts are complicated by: limited space investment needed to set up the process.
at virtually all sites; at different stages of the process, different  Work stoppage at any point ties up the whole
materials are needed; and the volume of materials needed is operation.
dynamic.  There is a lack of flexibility in handling a variety of
 3. (a) The advantages of a process layout are: products or production rates.
 It can simultaneously handle a wide variety of  6. Preconditions for high-volume, low-variety products are:
products or services, especially in terms of “batches” 1. Volume is adequate for high equipment utilization
or “job lots.” 2. Product demand is stable enough to justify high
 It has considerable flexibility with respect to investment in specialized equipment
equipment and labor assignments. 3. The product is standardized or approaching a phase of its
(b) The disadvantages of a process layout are: life cycle that justifies investment in specialized
equipment
 The use of general purpose rather than special purpose
4. Supplies of raw material and components are adequate
equipment tends to make the overall process
and of uniform quality to ensure that they will work with
somewhat less efficient.
the specialized equipment.
 Orders take more time and money to move through
the system because of the difficult scheduling, setting  7. Three types of work cells are the small work cell within a
up the process for a wide variety of orders, and plant, focused work center, and focused factory.
considerable material handling.  8. The advantages of work cells are:
 Labor skill requirements tend to be high because of  Reduction in work-in-process inventory
the use of general purpose equipment.  Reduction in required floor space
 Work-in-process inventories tend to be high.  Reduced raw material and finished goods inventory
 4. Most organizations have some procedure for documenting  Reduced direct labor cost
movement of their product. For instance, a hospital has doctor’s  Heightened sense of employee participation
orders indicating the tests and procedures that a patient is to  Increased utilization of equipment and machinery
undergo. A machine shop has routing documents indicating the  Reduced investment in machinery and equipment
operations that an order is to follow as the product moves through The disadvantages are:
the shop. And an auto repair shop knows what repairs, parts, and
 Similar to a product layout
labor are used on a particular job and as a result where the job was
 High volume is required because of the large investment
done and the trips made necessary.
needed to set up the process
In each of the above examples a matrix would be made
 There is a lack of flexibility in handling a variety of
showing the number of trips. And the distance (or time or cost) of
products or production rates
each trip would be determined.
 Requires the use of group technology
 5. (a) The advantages of a product layout are:  Requires a high level of training and flexibility on the part
 The use of special purpose equipment can make the of employees
overall process more efficient.

101
102 CHAPTER 9 L A Y O U T S T R A T E G I E S

 Either considerable staff support or imaginative

employees are needed for the initial development of the
work cells
 CHAPTER 9 L A Y O U T S T R A T E G I E S 103

 9. The requirements for a focused work center or focused factory Some points for class discussion:
are identification of a large family of similar products, a stable  Slotting fees mean that the small new company has
demand, and adequate volume to justify the capital investment. trouble getting products on the shelf.
10. Two major trends influencing office layout are dynamic  Slotting fees tend to reinforce the position of the strong/
needs for space/services, technology and virtual companies. major companies.
11. Some of the layout variables you might want to consider as  The retailer is interested in products that will sell and pro-
particularly important in an office where computer programs are vide margin—a strong inducement for any good product—
to be written are: if profitable, shelf space will be found.
 As a practical matter, slotting fees can take so many
 Ease of communication
forms (discounts, advertising fees, tasting stations, return
 Provision of privacy and a quiet work environment
policies, etc.) that real control is probably impossible.
 Lighting—especially as it related to glare on CRT
 Slotting fees are the inevitable result of the free market
screens
at work, one more item in the give and take between
 Consideration of ergonomic or human factor issues in
buyer and seller.
equipment layout and construction
 Slotting fees may make up half of the profit of a grocer.
12. Some drugstore and grocery chains now allow only limited  Since the fees are revenue for the retailers, the fees allow
routes through the store. This (a) allows a high traffic volume to be lower prices.
handled more readily, and (b) forces each customer to be exposed to  U.S. District Judge Gene Carter held in 2001, in Maine,
all merchandise. There are now Internet kiosks in shopping malls. that slotting fees in a news company distribution case
13. Retail store layout variables that a manager can manipulate were not illegal.
are: The authors see no problem with the ethics of slotting fees but
 Overall arrangement or pattern through the store would encourage firms not to hide the nature of the payments.
 The allocation of space to individual products
14. Each student will sketch the layout of a local supermarket. ACTIVE MODEL EXERCISE
They should observe the long aisles, power items at aisle caps,
and spread of staples at corners of store (fruit/meat/dairy/bakery). ACTIVE MODEL 9.1: Process Layout
15. “Random stocking” is placing a unit (product) anywhere in a 1. What is the current total cost?
warehouse that is open (available) for use. This is the opposite of $570
“static” storage where every SKU has its own allotted space. 2. Assembly (A) and Machine Shop (M) have the highest degree
16. Random stocking works best with sophisticated information of interaction. Would it be better to swap (A) and Painting or (M)
systems that rapidly identify items by bar codes (or other scannable and Painting in order to get (A) and (M) shop next to each other?
IDs), and place them randomly in a warehouse. That requires Swapping Assembly and Painting lowers the cost, while
(1) open location, (2) accurate records, (3) efficient “picking” swapping Assembly and Machine shop raises the cost.
sequencing, (4) combining orders, and (5) minimizing distance to 3. Use the Swap button one swap at a time. If the swap helps,
“pick” high usage items. move to the next pair. If not, click Swap to put the departments
17. Cross-docking means that units are never put in storage in back. What is the minimum total cost after all swaps have been
an intermediate storage facility (warehouse). Units are accurately tried?
labeled prior to shipment with the information needed to identify The cost drops to $510.
the final destination so that when they arrive at an intermediate 4. Look at the two data tables and use the yellow shaded column
warehouse or distribution point, they are immediately moved to to put processes in rooms. What room assignments lead to the
the shipping dock for transmittal to that final destination. minimum cost? What is this cost?
18. A heuristic is a “rule of thumb” method of problem solving. Putting the processes in rooms 1,4,2,5,6,3 respectively
Line balancing heuristics include: longest task time, most (yellow shaded area) has a minimum total cost of $460. (Other
following tasks, ranked positional weights, shortest task time, and layouts will have the same cost.)
least number of following tasks.
END-OF-CHAPTER PROBLEMS
ETHICAL DILEMMA 9.1 Movements = (4 × 8) + (9 × 7) + (7 × 4) + (6 × 3) + (8 × 2) + (10 × 6)
Slotting fees refer to payments by manufacturers for retail shelf
space—although the term can mean display or other support— = 32 + 63 + 28 + 18 + 16 + 60 = 217 (in 100s)
such as a tasting station—for a product. They are the result of: = 21,700
 New product proliferation (a grocer may handle only
Cost = 21,700 × $1 = $21,700
30,000 of 100,000 SKUs available, a book seller has a
similar problem)
 High rate of new product failure
 Increase in retailer power (the retailer as “channel
captain”)
104 CHAPTER 9 L A Y O U T S T R A T E G I E S

9.2 (a) Plan A movements  (20 × 6) + (12 × 18) + (8 × 2) Movement–Distance Calculations

 (6  4)  (10  2)  (4  18) M  W: 125  20 = 2,500
M  D:  75  40 = 3,000
= 120 + 216 + 16 + 24 + 20 + 72
M  G:  50  20 = 1,000
= 468 (in 100s) = 46,800 M  B:  60  20 = 1,200
W  L:  75  20 = 1,500
Cost = 46,800 × $0.50 = $23,400
D  G: 150  20 = 3,000
(b) Plan B movements = (20 × 6) + (8 × 18) + (12 × 2) D  B:  20  20 =  400
 (10  4)  (6  2)  (4  18) L  G:  20  20 =  400
13,000 = Minimum distance movement for
= 120 + 144 + 24 + 40 + 12 + 72 both (symmetrical layouts)
= 412 (in 100s) = 41,200 9.5
Cost = 41,200 × $0.50 = $20,600
Interstation Activity Matrix
(c) Plan C movements = (20 × 6) + (10 × 18) + (6 × 2) Pick Up Pick Up Verification
Paperwork Advising Class of Status
+ (8  4) + (12  2) + (4  18)
and Forms Station Cards and Payment
= 120 + 180 + 12 + 32 + 24 + 72 (A) (B) (C) (D)
Paperwork/   0 450 550  50
= 440 (in 100s)
 Forms (A)
= 44,000 Advising (B) 350   0 200   0
Class Cards (C)   0   0   0 750
Cost = 44,000 × $0.50 = $22,000 Verification/   0   0   0   0
(d) Plan B is the lowest cost, at $20,600.  Payment (D)
9.3 (23  10) + (32  5) + (20  8) = 230 + 160 + 160 = 550
Cost = 550  $2 = $1,100 Interstation Distance Matrix—Original Layout
9.4 This matrix includes movements in both directions: Pick Up Pick Up Verification
Paperwork Advising Class of Status
M W D L G B and Forms Station Cards and Payment
M — 125 75  0  50 60 (A) (B) (C) (D)
W — 0 75   0  0
Paperwork/  0 30 60 90
D —  0 150 20
 Forms (A)
L —  20  0
Advising (B) 30  0 30 60
G —  0
Class Cards (C) 60 30  0 30
B —
Verification/ 90 60 30  0
Here are two optimal layouts:
 Payment (D)

Load  Distance
A  B: 450  30 = 13,500
A  C: 550  60 = 33,000
A  D:  50  90 =  4,500
B  A: 350  30 = 10,500
B  C: 200  30 =  6,000
C  D: 750  30 = 22,500
90,000
or
Note: Work areas are “fixed” if using the software to reach this initial
answer.

(a) Initial layout:

 CHAPTER 9 L A Y O U T S T R A T E G I E S 105

Trip Matrix Distance Matrix

Interstation Distance Matrix—Improved Layout 1  2  3  4 5  1 2 3  4  5
Pick Up Pick Up Verification 1 0  8 13  0 0 1  0 7 8 12 14
Paperwork Advising Class of Status 2 5  0  3  3 8 2  7 0 5  6  7
and Forms Station Cards and Payment 3 3 12  0  4 0 3  8 5 0  4  9
(A) (B) (C) (D) 4 3  0  0  0 5 4 12 6 4  0  6
Paperwork/  0 30 30 60 5 0  8  4 10 0 5 14 7 9  6  0
 Forms (A)
Advising (B) 30  0 60 90 Tij  Dij = 602 (if areas are fixed; 566 if not; 595 if the sink is
Class Cards (C) 30 60  0 30 fixed in one location)
Verification/ 60 90 30  0 9.7 Layout 3:
 Payment (D)

Load  Distance
A  B: 450  30 = 13,500
A  C: 550  30 = 16,500
A  D:  50  60 =  3,000
B  A: 350  30 = 10,500
B  C: 200  60 = 12,000
C  D: 750  30 = 22,500
78,000 Trip Matrix Distance Matrix

(b) Improved layout: 1  2  3  4 5  1  2 3  4  5

1 0  8 13  0 0 1  0  4 8 12 14
2 5  0  3  3 8 2  4  0 4  8 10
3 3 12  0  4 0 3  8  4 0  2  8
4 3  0  0  0 5 4 12  8 2  0  4
5 0  8  4 10 0 5 14 10 8  4  0

Tij  Dij = 560 (if areas are fixed)

9.6 (a) Layout 1:
9.8 (a) Layout 4:

Trip Matrix Distance Matrix

1  2  3  4 5  1  2 3  4  5
1 0  8 13  0 0 1  0  4 8 12 16
2 5  0  3  3 8 2  4  0 4  8 12
3 3 12  0  4 0 3  8  4 0  4  8
4 3  0  0  0 5 4 12  8 4  0  4
5 0  8  4 10 0 5 16 12 8  4  0 Trip Matrix Distance Matrix Tij
 Dij
Tij  Dij = 600 with work areas fixed (504 if not fixed; 560 if 1  2  3  4 5  1  2 3  4  5
= 609
the sink is fixed in one location) 1 0  8 13  0 0 1  0  5 8 11 13
2 5  0  3  3 8 2  5  0 4  8 11
(b) Layout 2:
3 3 12  0  4 0 3  8  4 0  4  8
4 3  0  0  0 5 4 11  8 4  0  5
5 0  8  4 10 0 5 13 11 8  5  0
106 CHAPTER 9 L A Y O U T S T R A T E G I E S

(b) Layout 5:
9.12 Performance Time Task Must Follow
Trip Matrix Distance Matrix Task (in minutes) This Task
1  2  3  4  5  1 2  3  4  5 A  4 —
ref-1 0  8 13  0  0 ref-1  0 4 12 12  4 B  7 —
coun-2 5  0  3  3  8 coun-2  4 0  3  4  4 C  6 A, B
sink-3 3 12  0  4  0 sink-3 12 3  0  4  4 D  5 C
stor-4 3  0  0  0  5 stor-4 12 4  4  0  3 E  6 D
F  7 E
stov-5 0  8  4 10  0 stov-5  4 4  4  3  0
G  8 E
Tij  Dij = 478
H  6 F, G
Solution: Refrig. in Area 3 (where sink was); Counter in Area 4 49
(where storage was); Sink in Area 2 (where counter was); Storage
in Area 1 (where Refrig. was); Stove in Area 5 (no change) (a, d) 

9.9 B and C should be adjacent, because they have the most trips.
Traffic is next heaviest between A and D, so they should be
adjacent. Continuing in this fashion, F should be adjacent to D and
A should be next to F, but the latter two have already been placed.
Finally, E should be placed next to F. Thus, we are left with
(a) Heuristic solutions:
B C A D F E 47,900.
or
A D F E B C 44,440.
(Note: These are not the optimal solution.)
(b) Better layout:
A D F C B E 43,880.
9.10 (a) Takt time  Minutes available per day/Units demanded 480 minutes
per day (b) Cycle time   9.6 minutes
50 units
 420/250  1.68 minutes (c) Theoretical minimum   ti 49
  5.1  6 stations
(b) Number of cross-trained employees  (1.1  1.1  1.7 number of stations cycle time 9.6
 3.1  1.0)/1.68  8.0/1.68  4.76  5
Note that the theoretical minimum could not be achieved.
(c) The cleaning operation is substantially longer than the (e) Total idle time  4  1  2  3  2  1  2  15 min.
others so it warrants special consideration to ensure that per cycle  60 cycles  15 idle hours per day.
a smooth flow can be maintained. A machine constrained 49 minutes
task or lack of cross-training may suggest that more (f) Efficiency   0.766 or
8 stations  8.0 minutes*
traditional assembly line balancing techniques be used.
Efficiency 76.6%
9.11 (a) Cycle time = (60)(60 sec) *Longest operation time
180 PLAs
 task times
3,600 9.13 (a) Theoretical minimum number of stations =
= = 20 seconds per PLA cycle time
180
60
 task time Cycle time =  12 minutes. So minimum number of
(b) Theoretical minimum of workstations = 5
cycle time
48
60 stations =  4 stations
= =3 12
20
(c) Yes, it is feasible. Station 1 with A and C;
Station 2 with B and D; and Station 3 with E.
 CHAPTER 9 L A Y O U T S T R A T E G I E S 107

 60 min   60 sec 
(b) (b) cycle time =  40 hr  
 4,800 units   hr   min 
= 30 sec per unit
(c) t/CT 120/30 4 stations is theoretical minimum.
(d) The assembly line balance for a cycle time of 30 seconds
requires five stations, as shown below and in part (a)
above.

Time Time Left Ready

Station Task (sec) (sec) Tasks
A
1 A 20 10 B, C, D
This requires 5 stations—it cannot be done with 4. There is an
2 B 30 C, D
alternative solution with D and E in workstation #4 and F alone in
workstation #5. 3 C 15 15 D, E
D 15 E
48 48 4 E 10 20 F
(c) Efficiency    80% for 5 stations.
5 × 12 60 5 F 30 None
9.14
 (a)
Performance Time (e) Total idle time  30 seconds (10  20)
Task (in seconds) Predecessors
A 20 None
B 30 A (f) With five stations:
C 15 A t
Efficiency =
D 15 A (No. stations)(Cycle time)
E 10 B,C 120 120
F 30 D,E    .8 = 80%
(5)(30) 150
If six stations are used:
t
Efficiency =
(No. stations)(Cycle time)
120 120
   .666 = 66.6%
(6)(30) 180
9.15 (a)
108 CHAPTER 9 L A Y O U T S T R A T E G I E S

(b) Station 1 gets A, G, and B and has 0.5 minutes left 400 minutes
over. Station 2 gets C, D, and E, with no time left over. Cycle time 
60 units
Station 3 gets F, H, I, and J and has 0.5 minutes left
over. Improvements in efficiency would seem  6.67 minutes unit
impossible. The times are in 0.5 minute increments and  ti 25
Minimum number of stations  
can’t be sub-divided to achieve exact balance. Total of 1 Cycle time 6.67
minute of idle time/cycle.  3.75 or 4 workstations
(c) If stations 1 and 3 each had 0.5 minute more work to do,
the line would be 100% efficient; perhaps support tasks
could be assigned to them.

Time Time Left Ready

Station Task (minutes) (minutes) Tasks
1 A 5 5 B, G
G 3 2 B
B   1.5   0.5 C, E
2 C 3 7 E, D
D 4 3 E
Note: Four stations with a 7-minute cycle time is possible;
E 3 0 F Efficiency would become:
25
3 F 2 8 H  89.3%
H   3.5   4.5 I
4 7
I 2   2.5 J but production drops to 57 from the required 60 units. One
J 2   0.5 five-station solution (there are multiple answers) is:

(c) Theoretical:
Summary Statistics
Cycle time 10 minutes 25 minutes
Efficiency   0.937  93.7%
Time allocated (cyc  sta) 30 minutes/cycle 4 stations  6.67 minutes
Time needed (sum task) 29 minutes/unit Actual:
Idle time (allocated-needed) 1 minute/cycle
Efficiency (needed/allocated) 96.67% 25 minutes
Balance delay (1-efficiency) 3.333% Efficiency   0.833  83%
5 stations  6.00 minutes
(d) Idle time  1  1  1  2  5 min per cycle
(d) Theoretical minimum no. of stations  3
9.17 Performance Time Task Must Follow
9.16 Performance Time Task Must Follow Task (in minutes) This Task
Task (in minutes) This Task A  1 —
A  1 — B  1 A
B  3 A C  2 A
C  2 B D  1 C
D  4 B E  3 C
E  1   C, D F  1 C
F  3 A G  1   D, E, F
G  2 F H  2 B
H  5 G I  1   G, H
I  1   E, H 13
J  3 I
200
25 Cycle time   3.33min
60
(a, b)
(a) 
 CHAPTER 9 L A Y O U T S T R A T E G I E S 109

13 minutes
(b) Theoretical Efficiency   0.78
5 stations  3.33 minutes
or Efficiency 78% with a cycle time of 3.33. Multiple
layouts with this efficiency exist.
13
Actual efficiency   0.867 or 86.7%
5× 3
(c) Minimum number of  ti 13
   3.9  4
work stations Cycle time 3.33

(d) Idle time  1  1  2 min per cycle

 60 cycles (boats)  2 min or Efficiency  64.9%. Multiple layouts with this
 120 min; so it’s 2 hours per day efficiency exist.
13
Actual Efficiency   0.722 or 72.2%.
9.18 (a) Resolving Problem 9.17 with a production time of 300 36
minutes per day:
9.19 The assembly-line activities are:
300 minutes
Cycle time  Time Ready Time Station
60 units
Task (in minutes) Predecessors Tasks Left Assignment
 5 minutes unit
 ti 13 A  3 None A, B 1 1
Minimum number of stations   B  6 None C, D, E 1
Cycle time 5
C  7 A D, E, F 3 2
 2.6, or 3 workstations
D  5 A, B 3
E  2      B F, G, H, I 3 3
F  4      C 4
G  5      F J, H, I, K 2 4
H  7      D, E 5
I  1      H K, J 2 5
J  6      E 6
K  4 G, I, J 0 6
50

(a, g)

13 minutes
(b)  Efficiency   0.867
3 stations  5 minutes
or Efficiency 86.7%. Multiple layouts with this
efficiency exist.
(c) Resolving Problem 9.17 with a production time of 400
minutes per day:
400 minutes
Cycle time 
60 units
 6.67 minutes unit
 ti 13
Minimum number of stations  
Cycle time 6.675
 1.95 or 2 workstations
 24 hr   60min 
(b)   15min
 96 units   hr 

 1 unit   60 min   24 hr 
(c)  = 144 unitsper day.
 10 min   hr   day 
13 minutes
Efficiency   0.649
3 stations  6.67 minutes
110 CHAPTER 9 L A Y O U T S T R A T E G I E S

50 min per unit

(d)  5 stations Note that the efficiency of this particular grouping
10 min per cycle 28
  93.3%
Time needed per unit 56
(e) Efficiency 
Time allocated per unit Several other balances are also possible. One of them is
Total task time to place A alone, tasks B and C together, D and F together,
 E by itself, and G and H together.
(Cycle time) (Number of stations)
9.21 Note: This problem does not have a computer logo next to
50
 it, but it is possible for students to manipulate the input so that
(10)(6) POM for Windows or Excel OM can solve the problem.
50
 (a) One possible layout is:
60
 .8333, or 83.33%
(f) Idle time ime allocated per unit ime needed per
unit
60  50
10min/cycle
(g) Best assignment is shown in part (a) with the
efficiency shown in part (e) (i.e., 83.33%)
9.20
Performance Time Task Must Follow
Task (in minutes) This Task
A  5 —
B  3 A
C  4 B
D  3 B
E  6 C
F  1 C
G  4    D, E, F

H  2  G

28
 ti
Minimum number of stations 
Cycle time
28
  4 workstations
7
(b) The throughput is 3.75 patients per hour.
The work activities may be grouped, however, into no
fewer than five workstations without violating precedence (c) The bottleneck is at the medical exam station—16
requirements. minutes.
(d) Paramedics are idle 2 minutes, and doctors are idle 10
minutes for each patient.

28 28
Efficiency    0.80
5  7 35
or
Efficiency 80%
 CHAPTER 9 L A Y O U T S T R A T E G I E S 111

(e) If one more doctor and one more paramedic are added,
it is possible to increase the throughput to at least five Number of Number of
per hour with this simple layout: Task Successors Task Successors
A 9 F 2
B 4 G 2
C 4 H 1
D 2 I 1
E 2 J 0

Available
Available and Fit Assigned
Station 1 A A A
B, C — —
Station 2 B, C B, C C (Broke a tie)
B, F, G — —
Station 3 B, F, G B, F, G B
D, E, F G E, F, G F (Broke ties)
D, E, G — —
Station 4 D, E, G D, E, G D (Broke ties)
E, G E, G G (Broke a tie)
E, I — —
Station 5 E, I E, I E
I, H I, H H (Broke a tie)
I I I
J — —
Station 6 J J J
Answer: Station Tasks
(Other answers 1 A
The bottlenecks are now the two stations,
psychological exam and eye/measurements, taking 12 possible, depending 2 C
minutes. upon how ties are 3 B, F
broken in above 4 D, G
Σti 274 (seconds) procedure)
9.22 (a) n   5 E, H, I
Cycle time Cycle time (seconds) 6 J
60(60)seconds
Cycle time   60seconds (per iScan) (c) n = 6 workstations in our answer.
60 iScans
274 ti 274
so n   4.5667  n  5  Theoretical minimum (d) Efficiency =   0.7611
60 (No. Workstations)(Cycle time) 6(60)
number of workstations

(b) 60 seconds = cycle time 

 From (a)
Number of stations is at least 5

Precedence diagram:
112 CHAPTER 9 L A Y O U T S T R A T E G I E S

9.23 (a) Using a 40-hour week: Cottrell Bicycle Co.

Cottrell Bicycle Co. Longest Operation Time Cycle Time = 144 Seconds
Fewest Following Tasks  Cycle Time = 144 Seconds Station Task Time Time Left Ready Tasks
 1 K3  60  84 J3, K4, K9, J1, J2
Station Task Time Time Left Ready Tasks
J1  66  18 J3, K4, K9, J2
 1 J3 3 141 K3, F7 J3   3  15 K4, K9, J2, F7
F7 21 120 K3, C1  2 K9  27 117 K4, J2, F7, G5
C1 78  42 K3, B5 G5  29  88 K4, J2, F7
 2 B5 108  36 K3, A1, A2 K4  24  64 J2, F7, G4
 3 A2 72  72 K3, A1 J2  22  42 F7, G4, F3, F4
F3  32  10 F7, G4, F4, E2, E3
A1 52  20 K3
 3 E3 109  35 F7, G4, F4, E2
 4 K3 60  84 K4, K9, J1, J2
F7  21  14 G4, F4, E2, C1
K4 24  60 K9, J1, J2  4 F4  92  52 G4, E2, C1, D6
K9 27  33 J1, J2, G4 E2  18  34 G4, C1, D6
J2 22  11 J1, G4, F3, F4  5 G4  79  65 C1, D6, F9
 5 G4 79  65 J1, F3, F4, F9 D6  53  12 C1, F9, D8, D9
F3 32  33 J1, F4, F9, E3  6 F9 126  18 C1, D8, D9, D7
 6 F9 126  18 J1, F4, E3  7 C1  78  66 D8, D9, D7, B5
 7 E3 109  35 J1, F4 D9  37  29 D8, D7, B5
 8 J1 66  78 F4, G5  8 B5 108  36 D8, D7, A1, A2
 9 D8  78  66 D7, A1, A2
G5 29  49 F4, E2
A1  52  14 D7, A2
E2 18  31 F4, D7
10 D7  72  72 A2, B3
 9 D7 72  72 F4 A2  72   0 B3
10 F4 92  52 D6 11 B3  72  72 B7
11 D6 53  91 D8, D9 B7  18  54 A3
D8 78  13 D9 12 A3 114  30
12 D9 37 107 B3
Time allocated (cyc  sta) = 1728; Min (theoretical) # of stations = 11
B3 72  35 B7
Time needed (sum task)  1462; Efficiency 84.61%;
B7 18  17 A3
Idle time (alloc-needed)  266 seconds per cycle
13 A3 114  30
Cottrell Bicycle Co.
Time allocated (cyc  sta) 1872; Min (theoretical) # of stations 11
Most Following Tasks Cycle Time = 144 Seconds
Time needed (sum task) 1462; Efficiency 78.10%;
Idle time (allocated-needed) 410 seconds per cycle Station Task Time Time Left Ready Tasks
 1 K3  60 84 J3, K4, K9, J1, J2
J2  22 62 J3, K4, K9, J1, F3, F4
J3   3 59 K4, K9, J1, F3, F4, F7
K9  27 32 K4, J1, F3, F4, F7
F3  32 0 K4, J1, F4, F7, E3
 2 F7  21 123 K4, J1, F4, E3, C1
K4  24 99 J1, F4, E3, C1, G4
J1  66 33 F4, E3, C1, G4, G5
G5  29 4 F4, E3, C1, G4, E2
 3 F4  92 52 E3, C1, G4, E2, D6
E2  18 34 E3, C1, G4, D6
 4 C1  78 66 E3, G4, D6, B5
D6  53 13 E3, G4, B5, D9
 5 E3 109 35 G4, B5, D9, D8
 6 G4  79 65 B5, D9, D8, F9
D9  37 28 B5, D8, F9
 7 F9 126 18 B5, D8, D7
 8 B5 108 36 D8, D7, A1, A2
 9 D8  78 66 D7, A1, A2
A1  52 14 D7, A2
10 D7  72 72 A2, B3
B3  72 0 A2, B7
11 B7  18 126 A2, A3
A2  72 54 A3
12 A3 114 30
Time allocated (cyc  sta) = 1728; Min (theoretical) # of stations = 11
Time needed (sum task) = 1462; Efficiency = 84.61%;
Idle time (alloc-needed) 266 seconds per cycle
 CHAPTER 9 L A Y O U T S T R A T E G I E S 113

Cottrell Bicycle Co. Cottrell Bicycle Co.

Ranked Positional Weight Cycle Time = 144 Seconds Shortest Operation Time Cycle Time = 144 Seconds
Station Task Time Time Left Ready Tasks Station Task Time Time Left Ready Tasks
 1 K3  60  84 J3, K4, K9, J1, J2  1 J3 3 141 K3, F7
J2  22  62 J3, K4, K9, J1, F3, F4 F7 21 120 K3, C1
K9  27  35 J3, K4, J1, F3, F4 K3 60 60 C1, K4, K9, J1, J2
J3   3  32 K4, J1, F3, F4, F7 J2 22 38 C1, K4, K9, J1, F3, F4
F7  21  11 K4, J1, F3, F4, C1 K4 24 14 C1, K9, J1, F3, F4
 2 C1  78  66 K4, J1, F3, F4, B5  2 K9 27 117 C1, J1, F3, F4, G4
F3  32  34 K4, J1, F4, B5, E3 F3 32 85 C1, J1, F4, G4, E3
K4  24  10 J1, F4, B5, E3, G4 J1 66 19 C1, F4, G4, E3, G5
 3 G4  79  65 J1, F4, B5, E3, F9  3 G5 29 115 C1, F4, G4, E3, E2
 4 F4  92  52 J1, B5, E3, F9, D6 E2 18 97 C1, F4, G4, E3
 5 E3 109  35 J1, B5, F9, D6 C1 78 19 F4, G4, E3, B5
 6 F9 126  18 J1, B5, D6  4 G4 79 65 F4, E3, B5, F9
 7 J1  66  78 B5, D6, G5  5 F4 92 52 E3, B5, F9, D6
D6  53  25 B5, G5, D8, D9  6 D6 53 91 E3, B5, F9, D9
 8 B5 108  36 G5, D8, D9, A1, A2 D9 37 54 E3, B5, F9
G5  29   7 D8, D9, A1, A2, E2  7 B5 108 36 E3, F9, A1, A2
 9 E2  18 126 D8, D9, A1, A2, D7  8 A1 52 92 E3, F9, A2
D8  78  48 D9, A1, A2, D7 A2 72 20 E3, F9
D9  37  11 A1, A2, D7  9 E3 109 35 F9, D8
10 D7  72  72 A1, A2, B3 10 D8 78 66 F9
B3  72   0 A1, A2, B7 11 F9 126 18 D7
11 A1  52  92 A2, B7 12 D7 72 72 B3
B7  18  74 A2, A3 B3 72 0 B7
A2  72   2 A3 13 B7 18 126 A3
12 A3 114  30 A3 114 12

Time allocated (cyc  sta) = 1728; Min (theoretical) # of stations = 11 Time allocated (cyc  sta) = 1872; Min (theoretical) # of stations = 11
Time needed (sum task) = 1462; Efficiency = 84.61%; Time needed (sum task) = 1462; Efficiency = 78.10%;
Idle time (alloc-needed) = 266 seconds per cycle Idle time (alloc-needed) 410 seconds per cycle
Summary:
Number of
Algorithm Workstations Efficiency
Fewest following tasks 13 78.1
Longest operation time 12 84.6
Most following tasks 12 84.6
Ranked positional weight 12 84.6
Shortest operation time 13 78.1

The longest operation time, most following tasks, and

ranked positional weight algorithms gave equal efficiencies
of 84.61%.
114 CHAPTER 9 L A Y O U T S T R A T E G I E S

(b) Using a 41-hour week: Cottrell Bicycle Co.

Cottrell Bicycle Co. Longest Operation Time Cycle Time = 147.6 Seconds
Fewest Following Tasks Cycle Time = 147.6 Seconds Station Task Time Time Left Ready Tasks
Station Task Time Time Left Ready Tasks  1 K3  60  87.60 J3, K4, K9, J1, J2
J1  66  21.60 J3, K4, K9, J2
 1 J3   3 144.60 K3, F7
J3   3  18.60 K4, K9, J2, F7
F7  21 123.60 K3, C1
 2 K9  27 120.60 K4, J2, F7, G5
C1  78  45.60 K3, B5
G5  29  91.60 K4, J2, F7
 2 B5 108  39.60 K3, A1, A2
K4  24  67.60 J2, F7, G4
 3 A2  72  75.60 K3, A1
J2  22  45.60 F7, G4, F3, F4
A1  52  23.60 K3
F3  32  13.60 F7, G4, F4, E2, E3
 4 K3  60  87.60 K4, K9, J1, J2
 3 E3 109  38.60 F7, G4, F4, E2
K4  24  63.60 K9, J1, J2
F7  21  17.60 G4, F4, E2, C1
K9  27  36.60 J1, J2, G4
 4 F4  92  55.60 G4, E2, C1, D6
J2  22  14.60 J1, G4, F3, F4
D6  53   2.60 G4, E2, C1, D8, D9
 5 G4  79  68.60 J1, F3, F4, F9
 5 G4  79  68.60 E2, C1, D8, D9, F9
J1  66   2.60 F3, F4, F9, G5
D9  37  31.60 E2, C1, D8, F9
 6 F9 126  21.60 F3, F4, G5
E2  18  13.60 C1, D8, F9
 7 G5  29 118.60 F3, F4
 6 F9 126  21.60 C1, D8, D7
F4  92  26.60 F3, D6
 7 C1  78  69.60 D8, D7, B5
 8 D6  53  94.60 F3, D9
 8 B5 108  39.60 D8, D7, A1, A2
D9  37  57.60 F3
 9 D8  78  69.60 D7, A1, A2
F3  32  25.60 E2, E3
A1  52  17.60 D7, A2
E2  18   7.60 E3
10 D7  72  75.60 A2, B3
 9 E3 109  38.60 D7, D8
A2  72   3.60 B3
10 D7  72  75.60 D8
11 B3  72 75.60 B7
11 D8  78  69.60 B3
B7  18 57.60 A3
12 B3  72  75.60 B7
12 A3 114 33.60
B7  18  57.60 A3
13 A3 114  33.60 Time allocated (cyc  sta) = 1771.20; Min (theoretical) # of stations = 10
Time needed (sum task) = 1462.00; Efficiency = 82.54%;
Time allocated (cyc  sta) = 1918.80; Min (theoretical) # of stations = 10
Idle time (alloc-needed) = 309.20 seconds per cycle
Time needed (sum task) = 1462.00; Efficiency = 76.19%;
Idle time (alloc-needed) = 456.80 seconds per cycle
 CHAPTER 9 L A Y O U T S T R A T E G I E S 115

Cottrell Bicycle Co. Cottrell Bicycle Co.

Most Following Tasks Cycle Time = 147.6 Seconds Ranked Positional Weight Cycle Time = 147.6 Seconds
Station Task Time Time Left Ready Tasks Station Task Time Time Left Ready Tasks

 1 K3  60  87.60 J3, K4, K9, J1, J2  1 K3  60  87.60 J3, K4, K9, J1, J2
J2  22  65.60 J3, K4, K9, J1, F3, F4 J2  22  65.60 J3, K4, K9, J1, F3, F4
J3   3  62.60 K4, K9, J1, F3, F4, F7 K9  27  38.60 J3, K4, J1, F3, F4
K9  27  35.60 K4, J1, F3, F4, F7 J3   3  35.60 K4, J1, F3, F4, F7
F3  32   3.60 K4, J1, F4, F7, E3 F7  21  14.60 K4, J1, F3, F4, C1
 2 F7  21 126.60 K4, J1, F4, E3, C1  2 C1  78  69.60 K4, J1, F3, F4, B5
K4  24 102.60 J1, F4, E3, C1, G4 F3  32  37.60 K4, J1, F4, B5, E3
J1  66  36.60 F4, E3, C1, G4, G5 K4  24  13.60 J1, F4, B5, E3, G4
G5  29   7.60 F4, E3, C1, G4, E2  3 G4  79  68.60 J1, F4, B5, E3, F9
 3 F4  92  55.60 E3, C1, G4, E2, D6 J1  66   2.60 F4, B5, E3, F9, G5
D6  53   2.60 E3, C1, G4, E2, D9  4 F4  92  55.60 B5, E3, F9, G5, D6
 4 C1  78  69.60 E3, G4, E2, D9, B5 D6  53   2.60 B5, E3, F9, G5, D9
E2  18  51.60 E3, G4, D9, B5  5 E3 109  38.60 B5, F9, G5, D9, D8
D9  37  14.60 E3, G4, B5 G5  29   9.60 B5, F9, D9, D8, E2
 5 E3 109  38.60 G4, B5, D8  6 F9 126  21.60 B5, D9, D8, E2
 6 G4  79  68.60 B5, D8, F9 E2  18   3.60 B5, D9, D8, D7
 7 F9 126  21.60 B5, D8, D7  7 B5 108  39.60 D9, D8, D7, A1, A2
 8 B5 108  39.60 D8, D7, A1, A2 D9  37   2.60 D8, D7, A1, A2
 9 D8  78  69.60 D7, A1, A2  8 D8  78  69.60 D7, A1, A2
A1  52  17.60 D7, A2 A1  52  17.60 D7, A2
10 D7  72  75.60 A2, B3  9 D7  72  75.60 A2, B3
B3  72   3.60 A2, B7 B3  72   3.60 A2, B7
11 B7  18 129.60 A2, A3 10 B7  18 129.60 A2, A3
A2  72  57.60 A3 A3 114  15.60 A2
12 A3 114  33.60 11 A2  72  75.60

Time allocated (cyc  sta) = 1771.20; Min (theoretical) # of stations = 10 Time allocated (cyc  sta) = 1623.60; Min (theoretical) # of stations = 10
Time needed (sum task) 1462.00; Efficiency 82.54%; Time needed (sum task) = 1462.00; Efficiency = 90.05%;
Idle time (alloc-needed) 309.20 seconds per cycle Idle time (alloc-needed) = 161.60 seconds per cycle
116 CHAPTER 9 L A Y O U T S T R A T E G I E S

CASE STUDY
Cottrell Bicycle Co.
Shortest Operation Time Cycle Time = 147.6 Seconds STATE AUTOMOBILE LICENSE RENEWALS
Station Task Time Time Left Ready Tasks 1. What is the maximum number of applications per hour that
 1 J3   3 144.60 K3, F7 can be handled by the present configuration of the process?
F7  21 123.60 K3, C1 The process times and activities for each activity are identical be-
K3  60  63.60 C1, K4, K9, J1, J2 cause all have only one station. The maximum output of renewals
J2  22  41.60 C1, K4, K9, J1, F3, F4 will be limited to 60 renewals/hour (3600 sec/hr 60 sec/renewal)
K4  24  17.60 C1, K9, J1, F3, F4 by the bottleneck or longest process time.
 2 K9  27 120.60 C1, J1, F3, F4, G4 If each step in the process is handled by one person, it can be
F3  32  88.60 C1, J1, F4, G4, E3 seen that each station will be waiting for the clerk who checks the
J1  66  22.60 C1, F4, G4, E3, G5 file for violations. This is because this step takes the longest (60
 3 G5  29 118.60 C1, F4, G4, E3, E2 seconds). The task of “check file” will be the bottleneck, and a
E2  18 100.60 C1, F4, G4, E3 line will build up in front of this station. The clerk and expensive
C1  78  22.60 F4, G4, E3, B5 equipment for the photographic step will be idle approximately
 4 G4  79  68.60 F4, E3, B5, F9 1/3 of the time (20 seconds 60 seconds).
 5 F4  92  55.60 E3, B5, F9, D6 A balanced line process is one in which the process times of
D6  53   2.60 E3, B5, F9, D9 each station are the same. An obvious way to balance the line is to
 6 D9  37 110.60 E3, B5, F9 add stations to the bottleneck activity. However, this may not be the
B5 108   2.60 E3, F9, A1, A2 most efficient solution. In some cases, it is possible to combine
 7 A1  52  95.60 E3, F9, A2 activities creatively and make more productive use of workers.
A2  72  23.60 E3, F9 2. How many applications can be processed per hour if a
 8 E3 109  38.60 F9, D8 second clerk is added to check for violations?
 9 D8  78  69.60 F9 If a second file clerk is added to the activity of checking files,
10 F9 126  21.60 D7 the process time for this activity is reduced to 30 sec/location
11 D7  72  75.60 B3 (60 seconds/2 locations). The bottleneck now becomes the eye
B3  72   3.60 B7 test. The maximum output of renewals becomes 90 renewals/hour
12 B7  18 129.60 A3 (3600 sec/hour 40 sec/renewal).
A3 114  15.60 3. Assuming the addition of a second clerk, what is the
Time allocated (cyc  sta) = 1771.20; Min (theoretical) # of stations = 10 maximum number of applications the process can handle?
Time needed (sum task) = 1462.00; Efficiency = 82.54%; If activities 1, 2, and 3 can be successfully combined to form a
Idle time (alloc-needed) = 309.20 seconds per cycle new activity taking 105 seconds that is accomplished by the same
Summary: three people, the process time of the new combined task is 35
seconds. So without adding any personnel, as was done above, it
Number of
is possible to process up to 90 renewals per hour.
Algorithm Workstations Efficiency
Creative rearranging and combining of tasks can produce
Fewest following tasks 13 76.2 other cycle times. The only limitations are: (1) the tasks must be
Longest operation time 12 82.5 performed in a logical sequence, and (2) the facilities and equip -
Most following tasks 12 82.5 ment must be available for the tasks.
Ranked positional weight 11 90.1
4. How would you suggest modifying the process in order to
Shortest operation time 12 82.5
accommodate 120 applications per hour?
The ranked positional weight algorithm gave the highest This question requires trial-and-error creation of proposed
efficiency—90.05%. solutions. Presented below are proposed solutions that each result
in the capacity for handling 120 renewals per hour. Solution A
was achieved by simply expanding the number of stations
performing each job so that at least 120 licenses are processed per
hour. Solution B combines jobs such that the process time at most
stations equals the bottleneck process time (or cycle time).
Although this reduced the number of employees from 8 to 7, one
of these is an additional photographer with another camera. So the
total costs is increased. Solutions C and D produce the same costs
per renewal, $0.867, and both employ 7 persons.
Are solutions C and D equal? Some managers would argue
that the five people who each perform jobs 1, 2, and 3 have an
enriched job. Others would argue that enlarging a job is not the
same as enriching a job. How difficult will it be to monitor the
performance of each of these five people working independently
as a line process? How difficult will it be to teach each of the five
 CHAPTER 9 L A Y O U T S T R A T E G I E S 117

employees all four jobs rather than teaching each person one or VIDEO CASE STUDIES
two jobs. This is where the quantitative analysis ends and
judgment must be exercised. 1 LAYING OUT ARNOLD PALMER
HOSPITAL’S NEW FACILITY
A short video, available from Prentice Hall, was filmed specifically
Proposed Solutions—120 Renewals/Hour: for this text to supplement this case. (Running time is 9 minutes.)
Solution A A shorter (2-minute) version of the video also appears on the
student DVD in the text.
Maximum
Time Process Output 1. A hospital considers many variables in layout design. These
Job (sec) Station Time (sec) Cost/Hour include:
(per hr)
1 15 1 15    240 $12.00 a) Nursing efficiency: how far does a nurse have to travel in
2 30 1 30    120* $12.00 walking time and distance to patients, suppliers, break rooms,
3 60 2 30    120* $24.00 etc.
4 40 2 20    180 $24.00 b) Security: how to protect patients, especially babies from
5 20 1 20    180 16.00 10.00
being reached by people who should not have access to
6 30 1 30    120*   18.00
them.
Cost per renewal $116.00 120 $0.967 $116.00
* Indicates a bottleneck step. c) Privacy, serenity, and quiet: how to provide patients with
space that will help them recover quickly (servicescapes)
d) Meal and supply efficiencies: How to deliver meals warm
Solution B and keep supply lines full with the best efficiency.
Maximum e) Space for family to visit or sleep over.
Time Process f) Space for growth and capacity expansion.
Output
Job (sec) Station Time (sec) Cost/Hour
(per hr) g) Where to locate service departments (e.g., x-ray,
1 2 3 105 4 26.25  137.14    $48.00 pharmacy, laboratories, financial, management, etc.) so
4 5  60 2 30  120* 32.00 20.00 they are convenient to patients and staff.
6  30 1 30  120* 18.00
h) Areas for medical and non-medical staff to be “off-duty”
Cost per renewal $118.00 120 $0.983          $118.00
from patient/visitor flow.
* Indicates a bottleneck step. 2. The traditional “racetrack” or linear hallway layout is less in-
favor today than layouts that consider the efficiency of a limited and
Solution C expensive resource, i.e., nurses. The circular pod design makes it
much easier for nurses to see and reach each room. This not only
Maximum
saves travel time and walking or running exhaustion, but is more
Time Process Output
Job (sec) Station Time (sec) Cost/Hour
medically efficient when staff have a view of each room at all times.
(per hr)
1 2 3 + 4 145 5 29  124.1    $60.00 In the circular pod system, supplies for each two rooms are in a
5 20 1 20  180 16.00  10.00 cabinet at the pair’s entrances. This differs from a traditional
6 30 1 30   120* 18.00 layout with one supply room per nursing unit.
Cost per renewal $104.00 120 $0.867        $104.00 3. Using Figure 9.22’s linear layout, if nurse Smith makes 6
* Indicates a bottleneck step. round trips to each of the 12 patient rooms, 20 to medical supply, 5
to break room, and 12 to linen supply, his total distance traveled is:
Total distance = 6 trips  2 (for round trip)  (20 + 30 + 40 + 50 +
Solution D 60 + 70 + 20 + 30 + 40 + 50 + 60 + 70)′
Maximum + 20 trips medical supply  2  50′
Time Process
Output + 5 trips break room  2  40′
Job (sec) Station Time (sec)
(per hr) Cost/Hour + 12 trips linen room  2 30′
1+4 55 2 27.5   130    24.00 = 12(540)′ + 40(50)′ + 10(40)′ + 24(30)′
2 30 1 30   120*    12.00
= 6,480′ + 2,000′ + 400′ + 720′ = 9,600′ = 1.82 miles
3 60 2 30   120*    24.00
5 20 1 20  180 16.00 10.00 4. Travel Matrix for Figure 9.23 – Pod Design (Nurse Jones)
6 0 1 30   120* 18.00 Total = 7 trips  2(round-trip factor)  12 rooms  14′
Cost per renewal $104.00 120 $0.867         $104.00 + 20 trips to Central Medical Supply  2  60′
* Indicates a bottleneck step. + 6 trips to Break Room  2  60′
+ 12 trips to Pod Linen Supply  2  14′
= 2,352′ + 2,400′ + 720′ + 336′ = 5,808′ = 1.1 miles
118 CHAPTER 9 L A Y O U T S T R A T E G I E S

Note that Questions 3 and 4 did not use the same data, so we INTERNET CASE STUDY*
do not imply a reduction of this magnitude for these sample
MICROFIX INC.
numbers.
The initial analysis is straightforward. The line is balanced using
5. Servicescapes, a term coined by Professor Mary Jo Bitner in a the data exactly as presented in the case study. The theoretical
1992 Journal of Marketing article, deals with ambient conditions, minimum number of stations is 11. However, none of the rules
spatial layout, and signs/symbols/artifacts. we’ve looked at thus far leads to 11 stations. The following table
a) Ambient conditions, such as lighting, sound, and shows the answers given by the different rules:
temperature are all critical in Arnold Palmer’s neonatal unit,
Number of Minimum
as
Rule Stations Slack
well as the hospital as a whole. The case and video note
that the neonatal units have been completely redesigned, Longest operation time 13 0.16
with the result being a quicker recovery for premature Most following tasks 13 0.00
infants. Ranked positional weight 13 0.11
Shortest operation time 14 Irrelevant
b) Spatial layout, with its circular nursing pods, is the heart of Fewest following tasks 13 0.19
the new building.
The balance given by fewest following tasks is best, because
c) Signs/symbols/artifacts are very important in a hospital for it has the fewest stations and the cycle time can be reduced by the
children. A 30-foot-high genie hangs in the main Arnold largest amount of time. An alternative is to see what happens for
Palmer lobby. Art work, done by children when 41 hours (even though the case says not to). The following table
2 they were patients, is everywhere. Carpeting, colors, shows results:
play areas, and pictures are the artifacts that indicate this is
a warm and safe place for children and families. Number of Minimum
Rule Stations Slack
FACILITY LAYOUT AT WHEELED COACH Longest operation time 12 Irrelevant
This case study works best if the 7-minute video, made Most following tasks 12 Irrelevant
specifically for this text, is shown with the written case. Ranked positional weight 11 0.03
Alternatively, the student can view a 2-minute edited version of Shortest operation time 12 Irrelevant
the video on the book’s DVD. Fewest following tasks 13 Irrelevant

1. Assembly line balancing models, process layout to–from Clearly, the ranked positional weight balance leads to the
charts, and factory layout software can all be used in this case. fewest number of stations. Furthermore, the cycle time could be
2. The two major plants are across a busy street from one reduced (by 0.03) or the production rate could be increased by a
another, which has turned out to be detrimental to smooth product small amount without changing the number of stations.
flow. Likewise, to reach the warehouse, one must leave the main * This case study can be found at our Companion Web site, at www.prenhall.
structure. More preassembly of modular components in work cells com/heizer.
prior to the final 7-day assembly line would help.
3. The standard models for efficiency apply here. The firm
doesn’t want too much idle time at any workstation, and must
balance tasks very carefully.