Ristorazione 4.0

Ricerca e sviluppo di algoritmi di customer profiling, demand forecasting e decision support per l'ottimizzazione della gestione dei servizi di produzione e vendita del settore della ristorazione in ottica Industria 4.0


Development of Customer Profiling Algorithms


Development of Demand Forecasting Algorithms

Construction of a Decision Support System


Customer Profiling Data Analysis

Sales Forecasting Module

Order Schedule Form



monitored Key Performance Indicator


products to be discarded at the end of the time horizon


freshness product 

- 11%

unsold stocks quantity at the end of the time horizon


The object of study of the research conducted is the design of an integrated decision support platform for production and sales systems in the catering sector. The project aims to:

  • identify the decision support models applicable to the specific area;

  • define a decision support model for the catering sector from an Industry 4.0 perspective;

  • search and synthesize innovative algorithms of customer profiling and demand forecasting applied to the catering sector;

  • design, develop and test a decision-making engine for the optimization of planning and implementation of the "production on demand" functions of a restaurant system.

DSS: Decision Support System

Ristorazione 4.0: overview of R&D activities objectives

Idea75, with Ristorazione 4.0 has developed descriptive and predictive analytics algorithms for energy efficiency and predictive maintenance.
The decision support system aims to optimize production planning through demand forecasting systems that enable the implementation of production on demand.
This objective is achieved by designing a modular and reliable DSS, whose main constituents are the following:
  • the first realizes, starting from the data acquired, the sales forecasting which is used to determine the future demand; in this sense this module is dedicated to the automatic selection of the forecast model, based on some general criteria defined by the user;
  • the second provides support for order planning, including the multi-objective optimization method;
  • the third module performs a sensitivity analysis of the system in order to assess its performance and providing a Pareto front of proposals for optimal orders according to some Key Performance Indicators (KPIs) crucial for fresh and perishable products such as expiry dates, exhaustion of stocks and freshness.

DSS: sales forecasting module

Developed DSS structure

Stages of the customer profiling algorithm used

DSS: order schedule module


For the validation of Ristorazione 4.0, various test cases have been taken into consideration, different by product category; each of these has sales restrictions:

  • lot size (multiple orders of a minimum quantity);

  • delivery time;

  • execution times (when the order is issued).

Production planning based on sales forecasts was carried out by analyzing and monitoring various KPIs, such as:

  • Waste (waste): elements to be eliminated from the forecast time horizon due to the expiry of the shelf life;

  • Freshness: the age of the product when sold to the consumer;

  • Stock outs (out of stock): total demand not satisfied at the end of the forecast time horizon.

Furthermore, the out of samples indicators were evaluated for six different forecasting models and for the different optimization methods used.

The implemented decision support system dynamically adapts to the available data, selecting the most appropriate forecasting model based on criteria specified by the user (accuracy criterion or variability criterion).

Comparative tests were performed on the results obtained using the techniques implemented with respect to classical techniques used as benchmarks; the sensitivity analysis carried out states that there is always at least one model among those returned by the DSS that works better than the traditional ones.

With its technological and innovative solutions, Ristorazione 4.0 made it possible to obtain:

  • the reduction of production waste by 10%;

  • the increase in product freshness at the time of sale, estimated at 7%;

  • the reduction of the percentage of unsold stocks at the end of the time period, equal to 11%.

DSS: criteria selection for the most appropriate forecasting model

Segmentation / profiling system final architecture

Key Performance Indicator Analysis

Optimization methods used: comparison

Evaluation of error indices: comparison between traditional models and models validated by the DSS

stakeholders & Credits

Hs system


H.S. Systems offers business management software, analytics and cloud services to help organizations to improve performance and apply digital technologies to stop traditional thinking and enable new business models.


Ristorazione 4.0


Decision Support System

Quality Control

Demand forecasting

Customer Profiling




© 2019 by Idea75 srl. Proudly created with

  • Facebook Social Icon
  • Twitter Social Icon