Reducing Downtime with Predictive Analytics
Optimal Pricing for Alliance Trucking. Our client, Alliance Trucking has been serving both residential and commercial accounts in the greater Dallas / Fort Worth Metroplex since 1995. They are commonly contracted to provide delivery / removal services of all sorts of construction materials and waste.
Business Challenge
To precisely come-up with an algorithm that can accurately predict a customer acceptance quote for both the trucker and the material based on the historical data. The prediction should also consider the financials in such a manner that the quote should not only maximise the probability of acceptance by the given customer, but also should maximise the profit of the company.
Solution
Talentas have formulated a mathematical model based on the decision tree and random forest algorithm that considers the historical data to categorise the prominent explanatory variables based on each customer and job type. This mathematical equation is then subjected to iterative computation based on the mini-max principal to identify the optimum solution. That is, maximising the profit of the company by minimising the probability of rejecting the quote in an iterative manner.
Key Technologies & Platforms
- R Language for statistical computing
- Java based Web Application
Software Methodology
- Agile Iterative Development
Outcome
Talentas has developed an R code that is implanted to a web-based application in which the user can select the required artefacts and the system by itself will give the optimum results- which the user can directly sent to the customer as a formal quote for acceptance.