Presented by Group 5 | Section B PGDM (GM) 2015-16 Dhiraj Kumar [G15078] Naved Ahmed [G15091] Pankaj Kr. Goenka [G1509 Prashant Singh [G15100[ Ved Prakash [G15117] An Algorithm to Minimize transportation cost by Optimizing Vehicle type, Number of Trips & Part Loading for mass manufacturing industries Case : M/s Super Auto supplying to M/s Bajaj Auto, Chakan, Pune
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Presented by
Group 5 | Section B
PGDM (GM) 2015-16
Dhiraj Kumar [G15078]
Naved Ahmed [G15091]
Pankaj Kr. Goenka [G15096]
Prashant Singh [G15100[
Ved Prakash [G15117]
An Algorithm to
Minimize transportation cost by
Optimizing Vehicle type, Number of Trips & Part Loading for mass manufacturing industries
Case : M/s Super Auto supplying to M/s Bajaj Auto, Chakan, Pune
Background• M/s Super Auto (SAIL), Chakan supplies 10 parts to M/s Bajaj Auto, Chakan, Pune required for
production of various Pulsar / KTM motorcycles. Based on daily production plan the requirement of each of these parts varies from 0 to 3000 Nos.
• These parts are supplied in Reusable Pigeon Hole Plastic Bins of different sizes. Total there are 10 variety of Bins as Bin for each part is exclusive.
• For instance, Bin for Grab Handle holds 10 parts. So for a daily requirement of 200 parts, despatch representative would send 20 bins total. So on and so forth for all the parts.
• Currently to accommodate all the delivery requirement the despatch executive deploys the biggest permissible vehicle, Tata 909 – 2 or 3 in Nos. It is seen that these vehicles are not fully utilized resulting in cost loss.
• M/s SAIL has asked for a transportation cost increase and M/s BAL wants to suggest ways to M/s SAIL to improve vehicle planning & loading in order to minimize the cost. In process, it is required that all existing terms & conditions laid down by M/s BAL be followed.
Approach• Provide M/s BAL & M/s SAIL an easy to use spreadsheet to arrive at optimum number of trips
required for each vehicle type and total number of different Bins to be loaded onto that vehicle which would minimize the total transportation cost.
• Defining the current problem in terms of Input – Process – Output
Input Process Output
• BAL Daily Production Plan*• Part-wise total requirement in
Nos. (calculated)• Part Weight• Bin – LxBxH & Weight• No of Parts / Bin• Vehicle – LxBxH & Permissible
Wt.• Type of Permissible Vehicles
(T909/T709/T-Ace)*Daily/Weekly input, rest are one time inputs
• Define a Linear Program
• Convert Input parameters to a form that can be used in LP
• Execute LP
• Selection of Optimum Vehicle type out of 3 permissible variety
• Total number of trips required for each vehicle type
• Number of Bins of each part to be loaded in each trip
InputsDaily Plan (From Production Plan Sheet) Bin Dimension
Decision Variables Number of transportation trips for each vehicle type (3 Types of Vehicle) and Number of Bins of each of 10 parts to be loaded in each trip
Objective Function To minimize total transportation cost
Constraints• No production loss at BAL to be ensured.• Upto 5% excess supply can be done over daily plan.• No overloading (by Wt. or by Vol.) of trucks.• Only vehicle TATA 909, TATA 709 or TATA ACE or any combination of these three
vehicles can be used owing to dock constraints.
Process – Defining the LP
Decision Variables
xjyi – x1y1 would mean number of bins of part 1 loaded onto vehicle type 1. 30 such decision variables ( 10 parts x 3 vehicle types) ( j= 1 to 10 and i = 1 to 3)
Yi is the total number of trip of each vehicle type i 3 such decision variables
Objective Function
Min Z = Σ(Ci*Yi) where, Ci is the Cost per trip and Yi is the total number of trip of each vehicle type i, i=1 to 3, for Vehicle type 909, 407 & ACE respectively
Constraints
Weight Constraint
Σ(wj*xjyi) <= Wi*Yi where, xjyi is the total number of bins of type j loaded in vehicle type i. w j is the total wt. of each bin of type j. Wi is the permissible load wt of vehicle type i Total 3 constraints for i = 1 to 3, varying j=1 to 10 each time.
Process – Defining the LP
Constraints (contd…)
Volume Constraint
Σ(vj*xjyi) <= Vi*Yi where, xjyi is the total number of bins of type j loaded in vehicle type i. v j is the total vol. of each bin of type j. Vi is the permissible load volume of vehicle type i Total 3 constraints for i = 1 to 3, varying j=1 to 10 each time.
Total Supply Constraint (Production requirement must be met & tolerance)
Σ(xjyi) >= Total Min. no of Bins required for to be despatched for production – Total 10 constraints for j= 1 to 10, varying i=1 to 3 each time.
Σ(XjYi) <= Total Min. No of Bins required for production*1.05 – Total 10 constraints for j= 1 to 10, varying i=1 to 3 each time.
No of trips Constraint : Σ(Yi)<=100 (i=1 to 3)
Integer Constraint : Yi should be integer
Non-negative Constraint : All decision variables should be non-negative
Objective Function 3300.00(Minimize Transportation cost)
Trip Cost 1800 1500 1000 Truck Type y1 y2 y3 Decision Variable - No of trips of each type 1 1 0
x1 x2 x3 x4 x5 x6 x7 x8 x9 x10
Decision Variable and Constraints pertaining to Truck Type1
x1y1 x2y1 x3y1 x4y1 x5y1 x6y1 x7y1 x8y1 x9y1 x10y1 Decision Variable - No of Bins of each type 92 137 0 0 15 0 43 133 5 0 LHS RHS
Decision Variable and Constraints pertaining to Truck Type3
x1y3 x2y3 x3y3 x4y3 x5y3 x6y3 x7y3 x8y3 x9y3 x10y3 Decision Variable - No of Bins of each type 0 0 0 0 0 0 0 0 0 0 LHS RHS
Weight Constraint 7.84 11.15 5.68 11.02 15.35 11.67 6.23 5.33 11.77 11.89 0 <= 0Volume Constraint 0.012 0.0336 0.048 0.0336 0.048 0.0336 0.0096 0.048 0.0192 0.003 0 <= 0 Constaint pertaining to total Number of Bins of each type should be in required range
Total Number of Bins of each type 92 137 240 59 15 12 43 192 5 9
Before (Adhoc Planning) After (Optimized using LP)
Vehicle Type
Total Trips Cost Vehicle Type
Total Trips Cost
T909 2 3600 T909 1 1800
T709 - T709 1 1500
Ace - Ace -
Total per day
3600 Total per day
3300
Saving of 8% in transportation cost.
Scope of Horizontal Deployment
• The algorithm can be extended to all the vendors supplying parts in mass manufacturing setup for FMCG/ Automobile Industry
• Potential Saving of INR 3.5 Crs per annum if algorithm deployed horizontally at Bajaj Auto Ltd.
Total Part Procurement (Annual) (in Crs) 14000Assuming 25% needs optimization 3500Assuming 2% transport cost on part cost 70Assuming 5% improvement thru' LP 3.5
Scope of Improvement
• Scaling up algorithm for >25 parts and >5 vehicle variants