CASE STUDY 2 CHANDPUR ENTERPRISES LIMITED, STEEL DIVISION
Name:
Date:
SYED ABDUL RAHMAN BIN SYED AHAMED
816025
KHALIDUL ANWAR BIN ISHAK
818573
NOOR HANIM BINTI MD ISA
818933
10 JANUARY 2015
UUM CASE STUDY 2
SQQP 5023 – DECISION ANALYSIS
Table of Content
Chapter Title
Page
1.0
Introduction & Problem Statement
3
2.0
Analysis & Discussion
3-11
3.0
References
11
2
UUM CASE STUDY 2
SQQP 5023 – DECISION ANALYSIS
1. INTRODUCTION & PROBLEM STATEMENT Numerous administration choices include attempting to make the best utilization of a organization's assets. Assets commonly incorporate hardware, work, cash, time, warehouse space and crude materials. These assets might be utilized to make items such as hardware, furniture, sustenance or apparel or benefits, for an occasion, plan for aircrafts or creation, publicizing arrangements or venture choices.
Linear programming (LP) is a generally utilized scientific demonstrating method designed to help supervisors in arranging and choice making with respect to asset allocation. As talked about in Chandpur Enterprises Limited (CEL), Steel Division case study, and the organization overseeing executive needs to settle on the crude materials requirement for August creation at his steel plant.
Because of lower and upper limits on the measures of every crude material in a
batch
and changing measures of power and time devoured for distinctive crude materials, Akshay Mittal, overseeing chief of CEL can't just utilize the least expensive crude material. A linear program and Excel's Solver enhancement capacity will give the ideal amounts that meet the imperatives.
2. DISCUSSION & ANALYSIS 2.1
There is couple of vital focuses should be breaking down for better choice making which are;
a) What would be the best batch that could be making for one batch? b) What is the profit associated with this batch?
3
UUM CASE STUDY 2
SQQP 5023 – DECISION ANALYSIS
Decision variables: = kilograms of raw materials i to order per batch Related variables: fi = recovery * = finished goods tons of raw material i
The optimization is: Max [Revenue – Cost of RM – Electricity Cost – Consumables Cost – Salary Cost]
Where, per batch, Revenue = 29000 * ∑ /1000 Cost of RM = ∑ ( ∗ /1000 Electricity Cost = 4.30 * [700 *( ∑ )/1000 + 1200] Consumables Cost = 2000 * ∑ /1000 Salary Cost = 3000
a)
Constraint on batch size of 4,000 kg Figure 1: Solution to the batch model
4
UUM CASE STUDY 2
SQQP 5023 – DECISION ANALYSIS
So, to optimize a batch without any constraint related to monthly limits, profit per batch will be INR5, 421.
b)
Batch optimization with limits implied by monthly supply
Figure 2: Solution to a model with batch variables and linear limits implied by monthly supply
Alternative yields less per batch: INR 5,322. These shows yield more every month by doing more groups, 328 versus 321. There are more batches every month this optimization in light of the breaking point on the month to month supply. As a result of this constraint, Solver now becomes strength to utilize all the more excessive material rather than less expensive material. This enhances the general proficiency and in a roundabout way diminishes the time of one bunch.
5
UUM CASE STUDY 2
2.2
SQQP 5023 – DECISION ANALYSIS
Second analysis, will the administrative requirement of 4,000 kg for each batch of finished product hamper the capacity to make benefit? Is it worth to discover administrative endorsement to expand that point of confinement?
Ideal answers for LP have hitherto been discovered called, deterministic assumptions. Implies, presumption on complete assurance in information and relationship of a issue are characterize. On the other hand, conditions are continuing changing in certifiable just in this contextual analysis. Thus, to handle the error, significance of seeing just how touchy that arrangement is to model suspicions and information is essential.
Affectability examination only for the group without month to month requirements in view of this case study:
Figure 3: Sensitivity analysis for batch model without supply limits Final
Reduced
Objective
Allowable
Allowable
Value
Cost
Coefficient
Increase
Decrease
Tasla Raw Material per Batch (Kg)
1391.788450
0
2.67
0.72127846
0.56199723
Rangeen Raw Material per Batch (Kg)
1391.788450
0
3.37
1.E+30
1.00457297
Sponge Raw Material per Batch (Kg)
556.715379
0
2.14
0.56473868
6.98050847
Local Scrap Raw Material per Batch (Kg)
835.073069
0
2.37
0.65316276
4.65367232
0.000000
0
0.18
2.93138483
1.E+30
HC Raw Material per Batch (Kg)
1113.430760
0
1.24
1.E+30
0.49234907
Pig Iron Raw Material per Batch (Kg)
278.357690
0
2.24
0.80967742
13.9610169
Final
Shadow
Constraint
Allowable
Allowable
Value
Price
R.H. Side
Increase
Decrease
Tasla Raw Material per Batch (Kg)
1391.788450
0
0
1.E+30
1391.78845
Rangeen Raw Material per Batch (Kg)
1391.788450
1.03952679
0
1441.96107
1344.98991
Sponge Raw Material per Batch (Kg)
556.715379
0
0
1.E+30
2226.86152
Local Scrap Raw Material per Batch (Kg)
835.073069
`
0
1.E+30
3618.64997
0
0
0
1.E+30
4453.72303
HC Raw Material per Batch (Kg)
1113.430760
0.57320807
0
1751.31349
956.36581
Pig Iron Raw Material per Batch (Kg)
278.357690
0
0
1.E+30
278.35769
V ari able Cells
Name
Imported Scrap Raw Material per Batch (Kg)
Constraints
Imported Scrap Raw Material per Batch (Kg)
6
UUM CASE STUDY 2
SQQP 5023 – DECISION ANALYSIS
Tasla Raw Material per Batch (Kg)
1391.788450
0
0
1391.78845
1.E+30
Rangeen Raw Material per Batch (Kg)
1391.788450
0
0
1391.78845
1.E+30
Sponge Raw Material per Batch (Kg)
556.715379
-0.56395268
0
1386.96255
557.491289
Local Scrap Raw Material per Batch (Kg)
835.073069
-0.63952679
0
1344.98991
852.878465
0
-2.93138483
0
1331.55792
0
HC Raw Material per Batch (Kg)
1113.430760
0
0
1113.43076
1.E+30
Pig Iron Raw Material per Batch (Kg)
278.357690
-0.80347947
0
276.243094
280.504909
4000
3.39526792
4000
1.E+30
4000
Imported Scrap Raw Material per Batch (Kg)
Total Finished Product per batch (Kg)
Discussion: i.
Batch size is a big constraint on profits
ii.
If increase the batch size by 1 kg, profit increase per batch by INR3.40
iii.
If increase batch size by ~320 batches per month, profit increase to
~INR109, 000 (~6.25%) iv.
If it requires approximately INR1, 300,000 in capital and time investment to increase the batch size by just 100 kg, will able to recover that cost in less than 12 months
2.3
Third analysis, what amount of benefit will Akshay Mittal lose in the event that he should use in any event one unit of a crude material in a clump given or pick not to utilize that crude material? This is to stay away from miserable if CEL does not arrange a specific sort of crude material
From the sensitivity analysis in the case study:
i.
Row 13 indicates, Imported Scrap is the only raw material not being used in the current optimized plan which is the maximum profit per batch without any monthly limit constraint.
ii.
Row 31 shows, CEL would losing INR2.93 per additional kilogram if use Imported Scrap.
iii.
Suggest buying Imported Scrap if necessary and the price must below INR20, 070 per ton.
7
UUM CASE STUDY 2
2.4
SQQP 5023 – DECISION ANALYSIS
Forth analysis, Akshay Mittal must know the suggestions from ideal batch from question 2.1 on month to month commitment.
At the point when run Solver for boosting the benefit every month, benefit every month shows INR1, 788,705 which is much higher than the benefit every month assessed in question 2.1, INR1, 739,245. In the meantime, benefit per clump INR4, 873 dropped essentially from inquiry 2.1 INR1, 739,245.
Figure 4: Nonlinear model with batch decision variables and a monthly objective
8
UUM CASE STUDY 2
SQQP 5023 – DECISION ANALYSIS
The past methodology finishes up a shabby and ease crude material such as HC great in cluster plan and might incorporated in with the general mish-mash. Be that as it may, subsequent to this is a nonlinear model, there is probability that this enhancement may not produce a worldwide most extreme and only one of numerous nearby maxima. Along these lines, nonlinear model required to check if worldwide optima have. Nonlinear model need to use at many distinctive beginning stages to see dependably wind up at same ideal arrangement.
An approach to detail a straight month to month model is to utilize month to month crude material choice variables and include a choice variable for the quantity of batches. Month to month enhancement:
= tons of raw material i to order per month b = number of batches in a month Revenue = 29000 * ∑ Cost of RM = ∑ ( ∗ Electricity Cost = 4.30 * 700 *( ∑ ) + 1200 ∗ Consumables Cost = 2000 * ∑ Salary Cost = 3000 * b
Subject to min and max constraint for each i, constraint on batch size of 4,000 kg, batch size limit and hours available per month.
Monthly optimization: Max [Revenue – Cost of RM – Electricity Cost – Consumables Cost – Salary Cost]
9
UUM CASE STUDY 2
SQQP 5023 – DECISION ANALYSIS
Figure 5: Linear model with batch decision variables and a monthly objective
Discussion: i. ii.
2.5
Profit per month is same as profit per month for nonlinear monthly model. Nonlinear model did provide a global optimal
Last analysis, what are the suggestions to improve profits?
Based on sensitivity analysis (Fig 3): i. ii.
Find other sources for Rangeen to increase supply. Negotiate a deal with supplier and pay an amount up to an additional INR919 per ton of supply for each ton over the current limit of 500 tons.
10
UUM CASE STUDY 2
iii.
SQQP 5023 – DECISION ANALYSIS
Improve time per month from 600 hours to higher. Every one hour increase in time will result profit by INR2, 981. This additional profit would be applicable for the next 7.7 hours.
iv.
Improve the time per month: a. Hire better maintenance personnel to reduce maintenance time b. Use better / costlier machinery to reduce breakdown periods c. Timely supply of consumables and spare parts to reduce waiting time (emergency) d. Put in place a better safety plan for workers to reduce time in related activities.
3. REFERENCES i.
Render B., Stair, R.M., & Hanna, M.E. (2006). Quantitative Analysis for Management. Prentice Hall.
ii.
(2011), Chandpur Enterprises Limited, Steel Division: Teaching Note
11