Page 154 - Data Science Algorithms in a Week

138 Alfonso T. Sarmiento and Edgar Gutierrez

The solution yielded the results shown in Table 2. This table also includes parameters
a1, a2, a3, a4 which are the new equilibrium points for the state variables of interest.
Figure 5 shows the behavior of the state variables when this revised policy is applied.
The system has reached equilibrium approximately in 9 weeks (response time). This
figure also shows that the convergence of ADE has caused the asymptotic stability of the
four variables of interest. This was achieved mainly by increasing the parameter values
Desired Days Supply of Parts Inventory, Time to Correct Parts Inventory and Supplier
Delivery Delay and decreasing several other parameter values including Labor Recruiting
Delay, Preforms Cycle Time, and Presses Cycle Time.

Table 2. Parameter values for the stabilization policy

Parameter Value Unit
Desired Days Supply of Parts Inventory 3.46 Weeks
Time to Correct Parts Inventory 2.79 Weeks
Preforms Cycle Time 1.36 Weeks
Presses Cycle Time 1.70 Weeks
Time to Correct Inventory 1.47 Weeks
Supplier Delivery Delay 2.93 Weeks
Time to Adjust Labor 1.24 Weeks
Labor Recruiting Delay 0.5 Weeks
a1 (EP for Preforms WIP Level) 8828 Units
a2 (EP for Presses WIP Level) 13739 Units
a3 (EP for Finished Goods Inventory) 3275 Units
a4 (EP for Labor) 44 People

6,000

5,500

5,000

4,500

4,000
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
Time (week)
Maximum Capacity units/week

Figure 6. Maximum capacity of lenses manufacturing department.