Innovations such as mobile communications, prediction algorithms and machine learning have arrived in the multi-billion-dollar quality inspection and supplier compliance industry.
The impact will be transformational for the industry, which has been slow to adopt new technology. Companies (clients) that use inspection and supplier compliance services to inspect and evaluate suppliers’ production lines will benefit as well.
Traditional on-site supplier inspections are mired in outdated practices and limited analytical capabilities.
Client companies long ago outsourced inspection and supplier compliance, and today have limited or no access to the factory inspection process. This lack of transparency and accountability makes it very difficult to manage the quality of inspection services, making them prone to error and potential corruption. For instance, there is no guarantee that an on-site inspector working independently will sample products correctly or pay attention to problem areas that the client company has flagged.
Moreover, service providers and their customers need this type of information if they are to improve inspection practices and align them with the needs of different clients – which vary widely.
Typically, providers use digital cameras and spreadsheets – or even pen and paper – to record their findings. In addition to being painfully slow – it might take a day or two before the client sees the findings – these manual methods limit the scope for analyzing the data. Some services use online forms to record the results, which is an improvement on handwritten notes because aggregated data can be analyzed. But this method is still an inflexible and restrictive way to input inspection data and convey the findings to clients.
Yet the potential benefits of integrating today’s mobile communications platforms and analytics engines into the industry’s practices are significant. These changes are starting to emerge in the industry today.
One example in textile, apparel and footwear facility inspection is the hi-tech inspection system developed by Inspectorio. Major U.S. retailer Target recently appointed Inspectorio as a technology partner and the retailer is implementing the Inspectorio platform. The initial integration is in China and five other countries, and Target hopes to roll out implementation throughout all Target-contracted apparel manufacturing facilities by the end of the year.
In the Inspectorio system, when inspectors arrive at a factory they log into a mobile communications platform via their smartphones. These devices have all the capabilities that a modern inspection process requires: the ability to record visually and in audio, a geolocation feature, and plenty of data storage capacity.
All the information gathered by the application is recorded and stored in the cloud, and translated into a managerial dashboard that presents an overview of a factory’s performance.
The platform enables client companies to remotely monitor and participate in the inspection process in near real-time. This connectivity eliminates many of the risks associated with traditional practices, such as inspectors paying insufficient attention to specific areas of the production process. It provides documented and verifiable confirmation that an inspection took place in the manner specified by the client, with video and audio evidence. Also, the environment in which products are made can be visualized in real time, providing an immediacy that is absent from traditional inspection processes.
When the inspection is completed, a report can be generated electronically at the touch of a button. This is sent to the client company in real-time, as opposed to many the delays of hours or even days in the traditional scenario. The reports contain data on the inspector’s observations.
Beyond improving the quality and eliminating latency in the process, this innovative approach enables the use of prediction algorithms that can analyze historical data and identify common defects that might need special attention.
Machine learning capabilities continuously refine the inspection process. For example, when an inspection assignment is activated, an algorithm predicts the number and type of defects associated with the assignment based on historical performance data of the product category and the manufacturing location.
Using a similar algorithm, the platform can also improve the impact of compliance verification visits. The algorithm guides the auditor to high-risk areas in the manufacturing location, increasing the efficacy of each audit.
These are important advances, but this is only the beginning of the journey. Future inspection models could connect into more data sources and even more sophisticated analytical capabilities. As the models evolve, they may redefine the role of supplier quality and compliance monitoring services in global supply chains.
This post is based on the article written by Luis Moncayo (email@example.com), Co-Founder, Inspectorio, and James B. Rice, Jr., (firstname.lastname@example.org) Deputy Director, MIT CTL. The full article is published in the September/October 2017 issue of the journal Supply Chain Management Review.
Inspectorio is a member of the MIT CTL Supply Chain Exchange