# Pizza Delivery

This Market Simulation is designed to simulate the Market for Home Delivery Pizza in Square Town.

Square Town has 11 Streets (1st Street to 11th Street) and 11 Avenues (A-Avenue to K-Avenue). The population across Square Town is perfectly uniform, with the same number of people living on each block.

When the people of Square Town want a home delivered pizza, they have three options:

1. Downtown Dave
2. Suburb Sam
3. Online Olive

Downtown Dave has a single store located at the center of Square Town at the corner of 6th Street and K-Avenue. Suburb Sam has four stores – one located at each corner of the town (1st & A, 11th & A, 1st & K, 11th & K). Online Olive doesn’t have any storefronts and only accepts online orders.

For Downtown Dave and Suburb Sam, the delivery cost they charge customers is proportional to the distance between the nearest store and the customer – \$1.00 for every block traveled. For example, if Downtown Dave (at 6th & F) were to deliver a pizza to 4th & H then the delivery cost would be \$4.00 (the delivery truck would have to travel 2 blocks to 4th & F then another 2 blocks to 4th & H).

Online Olive has a different business model. Online Olive will deliver to any customer in Square Town for a fixed delivery fee of \$6.00.

This Case Study provides a high-level overview of the workflow without detailed explanation. It assumes you are already somewhat familiar with KNIME and Market Simulation. If not, start by reviewing the Building Blocks and Community Nodes.

# #1 Product Generator

The Product Generator Node will create an Output Willingness To Pay (WTP) Matrix for Customers from the set of Product Features and the Customer Distributions containing the part-worth values for each Feature.

For example, the first Customer in the Output WTP Matrix (Customer C00001) has a Willingness To Pay for Downtown Dave of \$12.98, for Suburb Sam of \$21.22, and for Online Olive of \$16.93. This makes sense as, we’ll learn later, that this Customer lives at 1st & A right next door to a Suburb Sam storefront. If the Price of all Pizza’s were the same then this Customer would buy from Suburb Sam.

## Input #2 Feature List

Optional – not used.

# #2 Geographic Feature

The Geographic Feature node calculates the ‘Lost Value’ of each Pizza option due to the delivery costs. Some Lost Value ‘Variation’ has been added to account for the fact that Customers care about more than just the cost of delivery – they also care about the time it takes for the pizza to be delivered. Hungry Customers are more willing to pay for rapid delivery.

The Geographic Feature node will also divide up the population so that it is distributed across the town. In this case, the population is evenly distributed. As there are 11 Streets and 11 Avenues, there are 121 intersections. As the town’s population is 10,000 there will be approximately 82 people located near each intersection. This can be seen in the Geographic Distributions Output.

# #3 Profit Engine

The Profit Engine takes the Input Product Array and the Input Willingness To Pay (WTP) Matrix can predicts how many Customers will buy pizza from each of the Stores. The Profit Engine node has been configured to also generate a Demand Curve for Downtown Dave.

The final Scatter Plot Charts show that the Customers who live around Downtown Dave will buy pizza from there, those in the outskirts of Square Town will buy from Suburb Sam, while those in between will buy from Online Olive. The Customers who decide to forgo buying any pizza are relatively evenly scattered throughout the town.