more in www.directingintelligence.com

 

 

1. Retail Network Performance Evaluation, new store best emplacement indication and profitability prediction

In retail business, it is crucial the ongoing performance evaluation of existing stores and the choice of the emplacement for a new one.

Based on historical data of existing stores (profitability, surface, employees, facilities, etc…), data concerning the social, demographic, economic and structural environment of each area, data about the competition and data concerning customers provided by Segmentation Application, we created DATACTIF RETAIL® Network Evaluator that realized with success the following tasks:

For new stores : Evaluation of new site location options, proposal for best emplacement and prediction of future profitability for each option.

For existing stores :

  • Profitability's Prediction for next years.
  • Estimation of the effect on the profitability in case of a new competitor appearance.
  • Estimation of the effect on the profitability in case that area properties change (metro station, commercial center, etc...).

 

2. Intelligent Retail Stock Management System

In the part Supplier _ Supermarket _ Consumer of the Supply chain, most important reason of food waste is the inefficient stock management into the Supermarket area.  

The other important reason is Customers demand. We have already created a model supported by a solution, DATACTIF RETAIL, that permits a deep understanding of consumption trends and we have also a consumption prediction model

Intelligent stock and waste management combining information from clusters, consumption prediction and indexes  such as waste factor and products expiration date, it performs stock optimization, products waste reduction, clients satisfaction.

 

3. Intelligent Service & Spare Parts Decision Support System

Using customers behavior toward service and data from the spare parts management system we created a Unified Classification -Prediction System based on spare parts usage, cost and criticality and customers attitude toward service & repairs with results :

  • Reduce spare parts management costs.
  • Particular emphasis on prediction for models whom parts production has stopped
  • Rating Official Dealers and support them in order to become more effective.
  • Identify and eliminate the usage of non-original spare parts  from some official dealers
  • Configuring policy after sales service in order to increase loyalty to official Dealers

 

4. Intelligent Telephony Network Optimizer

Based on a data combination such as Socio-Demographics, Relation History with Telephony Company, Call Detail Records (Number of calls for different services used such as : voice, SMS, data,  Duration of calls, Calling Behavior by Day/Time of the day for different call types, etc... ) and Geo -Spatial data, Intelligent Network Optimizer performs :

CLIENTS KNOWLEDGE AND PREDICTION

Deep understanding of customer behavior concerning mobile usage, mobility, social interaction, Economic Activity and

Prediction of future behavior, loyalty and churn, profitability, etc..

SERVICE QUALITY IMPROVEMENT

Understanding the dynamics of citizens’ daily mobility patterns is essential for the planning and management of urban facilities and services.  Knowing the moving patterns for different customer groups, a ser- vice provider can dynamically deploy resources to improve the service quality (e.g., adjusting the angles of antennas or re-positioning a mobile station).

One of the most promising and innovative ways to address the issue is to apply real-time big data analytics to constantly sense and optimize the Quality of Experience (QoE) of individual subscriber sessions under dynamically changing conditions. The necessary actions can then be applied automatically while the subscriber session is still live.

NETWORK OPTIMIZATION

During the last decades, with the growth of the Internet and cellular phones, there has been an increase in telecommunications demand. To support this growth, telecommunication companies are investing in new technologies to improve their services. In particular, real-time monitoring of infrastructures and services is a key issue within any telecommunication operator.

On one hand, the quality of service has to be assured by a timely fault diagnosis with evaluation of service impact and recovery. This is particularly needed to fulfill the Service Level Agreements (SLAs) that are set between the service provider and customers. On the other hand, operational tasks must be simplified to guarantee reduced Operating Expenses (OPEX).

An intelligent alarm management system will be capable of parsing the massive amount of received alarm events while reducing human intervention.

Each network generates sets of events that are related to the same situation. For example, in the access network of GSM systems, the Base Station Subsystem (BSS) contains Base Stations (BS) that are connected via a multiplexing transmission system to the Base Station Controller (BSC). These connections are very often realized with microwave line-of-sight radio transmission equipment.

Heavy rain or snow can temporarily disturb the connections between the antennas. But also massive movement of subscribers. The temporary loss of sight of a microwave disconnects all chained BS from the BSC and results in an alarm burst.

An intelligent alarm management will automatically discover these patterns classifying them by cause. When this information is combined with other features (e.g. trouble ticket data) it is possible to generate rules that lead to alarm correlation and root cause analysis