Proactive Employee Loss Prevention

Authors: 
Akshay Pandit, Prachi Chaudhari

Abstract:
This solution provides a smart validation system that safeguards shopper trust, faith, and loyalty towards the retailer for accuracy and reliability in store practices. This smart validation system is for Fixed lanes (manned by cashier) - using artificial intelligence (AI) and computer vision combined with real-time remote validation that is following the principle of ‘Monitor > Validate > Reconcile > Alert’ that verifies the items kept on the belt (by the shopper) is equal to the items scanned (by the cashier) is equal to the items bagged (by the cashier) is equal to the items paid (by the shopper)

Benefits –

1. Help detect theft and minimize shrinkage.
2. Boost associate efficiency & productivity.
3. Identify opportunities for training and behavior modification.
4. Uncover operational issues.
5. Monitor employee behavior and identify areas for performance improvement.
6. It detects scan avoidance, or cashier “sweet hearting,” bar code cover-up, direct bagging, item stacking, refund/return fraud detection, and even scanning two items at a time.

Background:
There are many factors that affect the profitability of the store. While some of these factors are outside of your control, loss prevention is not one of them. As a store manager, you have an obligation to minimize losses in your store. That may seem like a tall order, but when you and your employees follow procedures correctly, loss prevention will take care of itself.

Loss prevention is a series of strategies that minimize loss and protect your business from theft and shrinkage. Loss in retail stores affect everyone. When there are excessive losses, it affects the bottom line, the prices that must be charged to customers, and the money available to hire and pay staff members. There are many different causes of loss, but they almost all have one thing in common: people. Every member of your staff can play a role in creating loss — and preventing it.

Cashiers can prevent loss by staying focused when completing customer transactions and ensure the shopper is billed for the same number of items that the shopper placed on the belt. The shopper should not have to pay for the items that he/she did not intend to buy or which are not in his bag. When your employees’ performance is exceptional, your shrinkage continues to decrease while customer experience continues to increase.

Protecting the customer in such scenarios, which would help increase brand loyalty and customer satisfaction, is another key dimension to increasing sales and profitability.

Description:
  Persona 1 -

Shopper Sam, age 55, senior VP at a financial bank, tech savvy, always on his gadgets. A loyal customer and trusts his retailer.

  Situation –

Sam is heading home from work and on the way stops at his local retailer, he is multitasking, adding items in his cart while attending a conference call on his mobile. He places 10 items on the belt, his attention is still on his phone and lets the cashier complete scanning and bagging without interrupting.  He receives 10 items in the bag. Upon reaching home looks at the receipt and notices he is charged for 12 items instead of 10.

  Consequences –

1. Immediately Sam is forced to believe that this is not the first time and that he must be erroneously charged in the past transactions as well.
2. Sam accuses the retailer of theft or shoplifting committed by employees.
3. A retailer loses a loyal customer for life, which affects their existing pool of loyal customers.
4. Sam can spread the word of mouth and indulge in negative publicity on social media and in the local community which will tarnish the brand image of the retailer.
5. This act encourages some other employees with ill intentions to follow the same unethical practice.
6. Employee pocketing store supplies for personal use
7. Employee Selling or Upselling inventory for personal benefit.

Usages:

A smart validation system using artificial intelligence (AI) and computer vision combined with real-time remote validation that is following the principle of ‘Monitor > Validate > Reconcile > Alert’.

1. Monitor – The camera (on the belt) captures the information of the type and the number of items placed on the belt. Value: W
2. Validate –
   1. The camera (cashier overhead) and the scanning records capture the type and number if items getting scanned. Value: X
   2. The camera (in the bagging area) and the weight security mechanism capture the type and number of items getting bagged. Value: Y
3. Reconcile – The system stores the data for the type and number of items charged, i.e., number of items mentioned on the receipt. Value: Z. A comparison exercise is executed that validates W=X = Y = Z
4. Alert –
   1. If W= X = Y = Z then this record is saved in the log as ‘Success’
   2. In below example scenarios OR any other variance permutation, this record is saved in the log as ‘Discrepancy.’

Accordingly, appropriate action is defined to take counter measures. Here the transaction is not considered ‘Complete’ hence this is a proactive mechanism to prevent loss, as compared to manager intervention that could cause inconvenience and long queues.

Claims:
Claim is on the novelty of calculation in the usage section, currently data from individual components is not used to protect the customer, hence there is lot of human interventions around this use case, using the calculations derived from the data is not being used in the industry. Current loss prevention use cases are focused on customers. But not to protect the customer from theft happening in the store. With this calculation and implementation of it in assisted point of sale lanes, the idea is bridging that gap and protecting the customer.

Supporting Art:
Computer vision Product detection, Belt Sensors, Bagging area Camera’s.

TGCS Reference 4411