Provide a cost effective solution to allocate shipments to their nearest service centres. Each service centre has a defined delivery capacity.Lesser the distance of delivery location from service centre, more the cost effectiveness. Eg:
10,000 shipments to be delivered to end customers by 5 service centres in a city. 3 Service centres have a delivery capacity of 2000 shipments per day and 2 service centres have delivery capacity of 3000 shipments per day.
Factors to consider:
◉ Shipments have different dimensions and volumetric weight.
◉ Shipment delivery location can be equidistant from 2 service centers.
◉ Service centers use two wheelers for deliveries.
Our Solution Google Collab
Last mile delivery refers to the last leg of supply chain operations. A product's journey from a warehouse to the doorstep of the end-customer. This last step of the delivery process is most critical and should be optimized for a better user experience and dramatic reduction in operating costs.
We are proposing a Mixed Integer Programming (MIP) model for allocating shipments to the nearest service centers while keeping in mind the capacity constraints.
Further we formulate a Greedy heuristic and compare the performance of both the solutions.
MIP model :
Let number of warehouses be n and the number of shipments be m
cij be the cost of delivering a shipment from the ith Warehouse to the jth Shipment
location
Heuristic Greedy model :
◉ Initially assign each shipment to its nearest service center irrespective of the capacity
of the service center.
◉ Repeat the following steps until the number of shipments allocated to each service
center is less than it’s capacity.
Traverse over each service center and for each service center that has been over-loaded:
- Find the number of overloaded shipments = n.
- For each shipment assigned to this service center, we find the least possible
additional cost that will be incurred
if this shipment is allocated to another service center. - We transfer n shipments with the least transfer costs to their next nearest neighbour service center.
Allocation for Low density of Shipments
Allocation for High density of Shipments
Final Routes returned by the algorithm
Below are the steps to clone the repository and setting up the app in local.
- Python must be installed on your computer
-
Clone the Repository
git clone https://github.com/soul0101/Multi-Depot-Package-Allocation.git
-
Install Python packages that are used
pip install -r requirements.txt
-
Run main.py file
python main.py
