Capacity Planning

Problems

A manufacturing company has a product line consisting of five work stations in series. The individual workstation capacities are given. The actual output of the line is 500 units per shift.
Calculate (i) System capacity (ii) Efficiency of the production line

Workstation. No. Capacity/shift

A 600

B 650

C 650

D 550

E 600

(i) The capacity of the system is decided by the workstation with minimum capacity/shift, i.e., the bottleneck.
In the given example, the work station 'D' is having a capacity of 550 units/ shift which is a minimum. Therefore, the system capacity = 550 units/shift (ii) The actual output of the line = 500 units/shift. Therefore, the system efficiency = Actual output X 100 System capacity = (500) x 100 550 = 90.91 %

A company intends to buy a machine having a capacity to produce 1,70,000 good parts per annum. The machine constitutes a part of the total product line. The system efficiency of the product line is 85%. (i) (ii) Find the system capacity. The time required to produce each part is 100 seconds and the machine works for 2000 hours per year. If the utilisation of the machine is 60% and the efficiency of the machine is 90%, compute the output of the machine. Calculate the number of machines required?

(iii)

(i) System capacity =

Actual output / annum System efficiency

= 1,70,000 = 2,00,000 units / annum 0.85 = 2,00,000 = 100 units / hours 2,000
(ii) output per annum = Unit capacity x % utilisation x efficiency

Unit capacity = 60 x 60 sec = 36 units 100 sec per unit

output per hour = 36 x 0.6 x 0.9 = 19.44 units = 20 units (iii) Number of machines required = system capacity output per hour = 100 = 5 machines 20

The following activities constitute a work cycle. (i) Find the total time, (ii) Theoretical output obtained from the machine. (ii) Calculate the number of machines required to produce the three components from the information given below.

Sr. No.
(min) 1. 2. 3. 4. Components

Activity

Time

Unloading Inspection Loading job on machine table Machine operation time A 25 min 1.75 350 2450 B 55 min 3.0 550 4400

0.25 0.35 0.40 0.90 C 45 min 2.1 575 2875

1. Setup time per batch 2. Operation time (min/piece) 3. Batch size 4. Production per month

(i) Total cycle time (T) = 0.25 + 0.35 + 0.40 + 0.90 =1.90 min.
(ii) Output of the machine Output = 60 = 31.5 = 31 units. 1.9 (iii) Number of machines required Assume that the plant works on the single shift basis per day of 8 hours each. The total time required for the processing the components is given by Total time required = Setup time + operation time.
For component A, Total time required = Setup time + operation time = Production quantity x Setup time + Operation time Batch size Batch = [2450/350] x [25/60] + [2450 x 1.75/60] = 74.374 hrs.

For component A, Total time required = Setup time + operation time = Production quantity x Setup time + Operation time Batch size Batch = [2450/350] x [25/60] + [2450 x 1.75/60] = 74.374 hrs. For component B, Total time required

= [4400/550] x [55/60] + [4400 x 3/60] = 227.33 hrs.

For component C, Total time required

= [2875/575] x [45/60] + [2875 x 21/60] =104.375 hrs Total time (hrs) required to process' all the three components =74.374 + 227.33 + 104.375 = 406.079 hrs.

Total number of hours available (assuming 25 working days) per month =8 x 25 = 200 hrs.
Number of machines required = Total number of machine hours required Total number of hours available = 406.079 = 2.030 = 2 machines 200 Assuming a machine efficiency of 85% and operator efficiency of 75%, the number of machines required are: Total hours required per month = 406.079 = 636-98 hrs. 0.85 x 0.75 Number of machines required = 636.98 = 3.18 = 4 machines 200

Three components are to be manufactured on three machines i.e. Center lathe, Milling machine and Cylindrical grinding machine. (i) Calculate the number of machines required of each kind to produce the components if the plant works for 48 hours per week. (ii) Calculate the number of machines required assuming the machine efficiency of 75%. (iii) How do you reduce the number of machines. The following information is given: Machines 1.Center lathe 2.Milling machine Component A setup operation 30 min 2 min 45 min 8 min Component B setup operation 55 min 2.5 min 30 min 4 min 60 min 8 min Component C setup operation 40 min 1.5 min ---------------60 min 10 min

3.Cylindrical grinding 50 min 10 min Other details Lot size 350

400 4000

600 3000

Quantity demanded/month 1750

Demand Forecasting

Weighted Moving Average

The past data on the load on the weaving machines is shown below: a) b) Compute the load on the weaving machine centre using 5th moving average for the month of December 1996 B) Compute a weighted 3 month moving average for Dec 96 where the weights are 0.5 for the latet month,, 0.3 and 0.2 for the other months respectively

Month May 96

Load (Hrs) -

June
July Aug

585
610 675

Sep
Oct Nov

750
860 970

a) Five months moving average forecast for Dec 96 = (DNov+DOct+Dsept+DAug+DJuly)/5 = (970+860+750+675+610)/5 = 773 hrs. b) A three months weighted moving average forecast for Dec96 = (WNov.x DNov) + (WOct x DOct) + (WSept x DSept) =(0.5 x 970) + (0.3 x 860) + (0.2 x 750)

= 947.8 hrs

Estimate the sales forecast for the year 2000, using exponential smoothing forecaster. Take =0.5 and the forecast for the year 1995 as 160 x 105 units. Compare the forecast with least square method.

Year Sales Rs (x105)

1995 180

1996 168

1997 159

1998 170

1999 188

Sol.: The demand for the year 2000 can be computed using least square method as follows:

Year 1995 1996 1997

Sales (x 105) 180 168 159

Deviation (x) -2 -1 0

x2 4 1 0

xy -360 -168 0

1998
1999 N=5

170
188 y=505

+1
+2 x=0

1
4 x2 =10

170
376 xy=18

Substituting these values in the equations

a=y/N = 505/5 = 101 b=xy/ x2 =18/10 = 1.8

The regression equation is y=100 + 1.8x The demand for year 2000 = Rs.164.4 x 105

The forecast demand using the smoothing constant =0.5 is computed as shown in the table using the formula Ft = .Dt-1 + (1- ).Ft-1

Period 1995 1996 1997

Demand 180 168 159

Forecast demand for =0.5 160 (Given) 0.5 x 180 + (1-0.5) x 160 0.5 x 168 + (1-0.5) x 170

Ft

170.0 169.0

1998
1999

170
188

0.5 x 159 + (1-0.5) x 169

0.5 x 170 + (1-0.5) x 164

164.0
167.0

2000

0.5 x 188 + (1-0.5) x 167

177.5

The forecast demand for the year 2000 with =0.5 is 177.5

A dealer for electrical appliances forecasts demand for the Geyser at the rate of 500 per month for the next three months. The actual demands turned out to be 400, 560 and 700. Calculate the forecast error and bias, comment on the same Solution: MAD =n j=1 (Forecasted demand Actual demand) /3 = [(500 400) + (500 560) + (500 700)]/3 = (100 + 60 + 200)/3 = 120 units Bias = [(500-400) + (500 560) + (500 700)]/3

= [(100 60 200)]/3
= -0.53 units In this case MAD is 120 units and bias is -0.53 units Since MAD increases absolute error, we can conclude that it has very high degree of average error i.e 24% of the forecasted value of Geysers The Bias The forecasting method has a tendency to under estimate the forecast since bias is ve. i.e it is about 9.6% under forecast.

Q.1 Given below are the details of a fan assembly line. Suppose we want to assemble 750 fans per day. What would our cycle time have to be? Draw the Precedence Diagram. What is the theoretical minimum number of workstations for this problem? Find the efficiency of this line. Suppose the demand rose from 750 to 1000 units per day. What would you do? Task A B C D E F G H Description Time (sec) Assemble Frame 20 Mount switch 7 Assemble motor housing 20 Mount motor housing in frame 22 Attach blade 15 Assemble and Attach safety grill 10 Attach cord 16 Test 8 Predecessor(s) none A B B C D E, F G

Example of Job Sequencing: First-Come First-Served

Suppose you have the four jobs to the right arrive for processing on one machine.
What is the FCFS schedule? Do all the jobs get done on time?
Jobs (in order of arrival) A B C D Processing Time (days) 4 7 3 1 Due Date (days hence) 5 10 6 4

Example of Job Sequencing: First-Come First-Served

Suppose you have the four jobs to the right arrive for processing on one machine.
What is the FCFS schedule? Do all the jobs get done on time?
Jobs (in order of arrival) A B C D Processing Time (days) 4 7 3 1 Due Date (days hence) 5 10 6 4

Answer: FCFS Schedule

Jobs (in order of arrival) A B C D Processing Time (days) 4 7 3 1 Due Date Flow Time (days hence) (days) 5 4 10 11 6 14 4 15

No, Jobs B, C, and D are going to be late.