upadhyan

import numpy as np
import gurobipy as gp
from gurobipy import GRB

## First Stage Constants
i = np.array([6,14,27])
s = np.array([1,5,3])
B = 1000
S = 100
## Second Stage Constants

upadhyan

# SOLVING
m.optimize()
result = x.x

print(f"Our Server farm will cost ${np.round(m.ObjVal,2)}. We should buy {result[0]} CPUs, {result[1]} GPUs, and {result[2]} TPUs ")
# Our Server farm will cost $995.6. We should buy 51.0 CPUs, 5.0 GPUs, and 8.0 TPUs

upadhyan

# CONSTRAINTS 
## Installation Budget
m.addConstr(i @ x <= B)

## Space
m.addConstr(s @ x <= S)

## Meet demand in each scenario
m.addConstrs((
    x + y[:,k] >= d_xi[:,k] for k in range(K)

upadhyan

# OBJECTIVE 
m.setObjective(i @ x + gp.quicksum(c @ y[:,k] * p[k] for k in range(K)))

upadhyan

m = gp.Model("2SP")
m.ModelSense = GRB.MINIMIZE
m.setParam('OutputFlag', 0)  # Telling gurobi to not be verbose
m.params.logtoconsole=0
# VARIABLES 
## First Stage Decision variable
x = m.addMVar((3,), vtype = GRB.INTEGER, name = "x")

## 2nd stage decision for each potential scenario
y = m.addMVar((3,K), vtype = GRB.INTEGER, name = "y")

upadhyan

import numpy as np
import gurobipy as gp
from gurobipy import GRB

## First Stage Constants
i = np.array([6,14,27])
s = np.array([1,5,3])
B = 1000
S = 100
## Second Stage Constants