This project has as protagonist one of the most important world players in the construction of large chemical and petrochemical plants. With Moxoff, he developed a predictor, based on mathematical models, able to estimate the types and quantities of materials needed for a new project.
Description and Benefits
Before the intervention of Moxoff, the company made estimates based exclusively on the experience of its technicians, with a great use of time and resources. Relying on the availability of historical data and the in-depth knowledge of the company’s experts, Moxoff has developed an algorithm that returns operators a preliminary estimate of the materials to be provided to the design engineers.
The algorithm allows the company not only to calculate the total quantities of materials needed for the project, but is able to provide details on the different parts and types of material of which the overall plant is composed.
Augmented human: the encounter between mathematical models and human experience
The estimate made for this customer is a set of models that estimates the various types of materials (pipes, valves, flanges, etc.). Before building an overall final output, the software provides operators with the opportunity to check the quantities step by step and make any corrections, based on their experience, to obtain an increasingly accurate result. It is also possible for the user to provide guidance to the model on expected input-output relationships to guide the estimation process in the most unexplored contexts. These results are then reworked by our models to arrive at the final overall estimate.
- Quick estimates, decrease in processing time (from days and weeks to hours)
- More accurate planning, quantities and types of materials needed for the project can be known in advance
- A more accurate estimate, there is less chance that an estimate will deviate too much from the actual cost
