Method, Medium, and System for Demand Planning
A system and method are disclosed including a demand planner that receives a demand for two or more options that are needed to produce at least one automobile. The demand planner also models the two or more options as a network of arcs and nodes and generates one or more valid configurations of the two or more options. The demand planner further determines the demand for the one or more valid configurations and causes at least one manufacturer to manufacture, the at least one automobile based on the determined demand for the one or more valid configurations.
1 . A computer-implemented method for generating, solving and modifying a network graph, comprising:
modeling, by a computer comprising a processor and a memory, two or more options as a network of arcs and nodes, wherein the nodes range from 35,000 nodes to 2.66×10 6 nodes;
selecting, by the computer, one or more nodes on the network and receiving one or more option rules and constraints corresponding to each of the one or more nodes;
pruning, by the computer, the selected one or more nodes according to the one or more option rules and constraints;
rearranging, by the computer, the network model after the pruning to eliminate redundant arcs and nodes;
validating, by the computer, any arc added to the network model according to the received constraints and removing any invalid arc; and
generating, by the computer, a linear programming optimization model from the validated network model by associating a vector of binary values with the validated network model.
2 . The method of claim 1 , wherein a demand take rate of the network model is based on data comprising one or more of: past sales, a market of one or more countries, a population, a dealer and a manufacturer.
3 . The method of claim 1 , wherein the rearranging the network model comprises incorporating one or more logical substitutions into the network model.
4 . The method of claim 1 , further comprising:
generating, by the computer, equations representing flow conservation in the validated network model and representing demand take rate constraints comprising slack and surplus variables.
5 . The method of claim 1 , wherein the network model comprises a decision variable associated with each arc.
6 . The method of claim 1 , wherein a decision variable of an arc of the network model represents a percentage or an amount of configurations associated with the arc.
7 . The method of claim 1 , further comprising:
adding, by the computer, slack and surplus variables to one or more decision variables.
8 . A non-transitory computer-readable medium comprising software for generating, solving and modifying a network graph, the software when executed configured to:
model two or more options as a network of arcs and nodes, wherein the nodes range from 35,000 nodes to 2.66×10 6 nodes;
select one or more nodes on the network and receiving one or more option rules and constraints corresponding to each of the one or more nodes;
prune the selected one or more nodes according to the one or more option rules and constraints;
rearrange the network model after the pruning to eliminate redundant arcs and nodes;
validate any arc added to the network model according to the received constraints and removing any invalid arc; and
generate a linear programming optimization model from the validated network model by associating a vector of binary values with the validated network model.
9 . The non-transitory computer-readable medium of claim 8 , wherein a demand take rate of the network model is based on data comprising one or more of: past sales, a market of one or more countries, a population, a dealer and a manufacturer.
10 . The non-transitory computer-readable medium of claim 8 , wherein the rearranging the network model comprises incorporating one or more logical substitutions into the network model.
11 . The non-transitory computer-readable medium of claim 8 , wherein the software when executed is further configured to:
generate equations representing flow conservation in the validated network model and representing demand take rate constraints comprising slack and surplus variables.
12 . The non-transitory computer-readable medium of claim 9 , wherein the network model comprises a decision variable associated with each arc.
13 . The non-transitory computer-readable medium of claim 12 , wherein a decision variable of an arc of the network model represents a percentage or an amount of configurations associated with the arc.
14 . The non-transitory computer-readable medium of claim 11 , wherein the software is further configured to:
add slack and surplus variables to one or more decision variables.
15 . A system for generating, solving and modifying a network graph, comprising:
a computer comprising a processor and a memory, the computer configured to:
model two or more options as a network of arcs and nodes, wherein the nodes range from 35,000 nodes to 2.66×10 6 nodes;
select one or more nodes on the network and receiving one or more option rules and constraints corresponding to each of the one or more nodes;
prune the selected one or more nodes according to the one or more option rules and constraints;
rearrange the network model after the pruning to eliminate redundant arcs and nodes;
validate any arc added to the network model according to the received constraints and removing any invalid arc; and
generate a linear programming optimization model from the validated network model by associating a vector of binary values with the validated network model.
16 . The system of claim 15 , wherein a demand take rate of the network model is based on data comprising one or more of: past sales, a market of one or more countries, a population, a dealer and a manufacturer.
17 . The system of claim 15 , wherein the rearranging the network model comprises incorporating one or more logical substitutions into the network model.
18 . The system of claim 15 , wherein computer is further configured to:
generate equations representing flow conservation in the validated network model and representing demand take rate constraints comprising slack and surplus variables.
19 . The system of claim 15 , wherein the network model comprises a decision variable associated with each arc.
20 . The system of claim 18 , wherein a decision variable of an arc of the network model represents a percentage or an amount of configurations associated with the arc.