Sensor-Based Product Interaction Detection for Targeted Advertising

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing advertising methods fail to provide personalized and targeted advertisements effectively, as they lack the ability to accurately determine customer interest levels in products displayed on shelves, leading to irrelevant advertisements and inefficient advertising campaigns.

Innovation Solution

A system that uses sensors attached to products on display to acquire data, which is wirelessly transmitted to receivers and processed by a computer executing machine learning and localization algorithms to determine customer interactions and generate targeted advertisements based on insights about product interactions, including interaction time and location.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional advertising methods are used, then advertising coverage is broad, but advertising relevance to customer interests is low

Engineering Contradiction:

Improvecustomer interest level determination accuracyVSAvoidcustomer interaction data

Core Design Contradiction:

Measurement precisionVSLoss of information

Solution Approach 1:

Sensors are attached to products in advance to collect interaction data before advertising delivery, enabling proactive customer interest assessment. The system preliminarily gathers data on customer interactions (pickup duration, repeated picks) and uses this information to determine interest levels before serving targeted advertisements, resolving the contradiction by preparing measurement capabilities beforehand.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent replaces traditional mechanical survey methods (customer feedback forms, surveys) with automated sensor-based detection systems. Sensors objectively measure interaction parameters (pickup duration, frequency) and machine learning algorithms automatically determine interest levels, substituting subjective mechanical processes with objective electronic measurement and analysis systems.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If sensors are attached to all products, then customer interaction data collection is comprehensive, but system complexity increases

Engineering Contradiction:

Improveinteraction detection accuracyVSAvoidsensor network complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The system segments the product display area into multiple zones with sensors strategically placed on representative products within each segment. Rather than requiring sensors on every single product, the system divides the space and uses sensors on selected products to infer customer behavior patterns across similar product categories, reducing overall sensor quantity while maintaining measurement precision.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The sensor system is designed with multi-functionality to reduce complexity. The same sensor infrastructure serves multiple purposes: detecting product pickup events, measuring interaction duration, tracking repeated interactions, and providing localization data. This universal sensor platform eliminates the need for separate systems for each measurement function, thereby reducing overall device complexity.

Inventive Principle:

Principle #6Universality (Multi-functionality)

3Speed

If real-time advertising is provided, then advertising timeliness is improved, but processing requirements increase

Engineering Contradiction:

Improveadvertising delivery speedVSAvoidcomputational power consumption

Core Design Contradiction:

SpeedVSPower

Solution Approach 1:

The system applies partial processing by focusing computational resources only on products with detected customer interactions. Rather than continuously processing data for all products, the system activates detailed analysis and advertising generation only when sensor data indicates a pickup event, using excessive action (full processing) only when necessary and partial action (minimal processing) for products without interactions, thereby balancing speed and power consumption.

Inventive Principle:

Principle #16Partial or excessive action

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

Improves the accuracy of determining customer interest levels, providing more relevant advertisements, enhancing the effectiveness of advertising campaigns and improving the shopping experience, thereby increasing revenue for retailers and advertisers.

Implementation Method 1

acquiring, by one or more sensors attached to a product on display in a product display area, sensor data

Methodology Applied

Scientific EffectSensor detection:

Implementation Method 2

wirelessly transmitting the sensor data to at least one wireless receiver located in the vicinity of the product display area

Methodology Applied

Scientific EffectWireless transmission:

Data Source

PatentUS11017434B2Interactive product display system for providing targeted advertisements

Publication Date: 2021.05.25 NEC CORP

AI summary

A method for generating data indicative of a customer interaction with a product on display includes acquiring, by one or more sensors attached to a product on display in a product display area, sensor data; wirelessly transmitting the sensor data to at least one wireless receiver located in the vicinity of the product display area; and transmitting the sensor data from the at least one wireless receiver to a computer. The method further includes executing, by the computer, a machine learning algorithm in order to determine that a potential customer has interacted with the product on display and to generate insights about an interaction between the customer and the product on display.