9 minute read 1 Sep 2020
Container ship in import export and business logistic

How natural language processing can build supply chain resiliency

By Sven Dharmani

EY Global Advanced Manufacturing & Mobility Supply Chain Leader

Passionate about transforming supply chains. Problem solver. Curious and collaborative. Avid traveler, scuba diver and car enthusiast.

9 minute read 1 Sep 2020

Use NLP to monitor social media and other content for data on disruptive events to brace your supply chains sooner, rather than later.

In brief
  • Since most companies use internal data to track demand-supply balances, it’s challenging for them to foresee external events impacting their supply chains.
  • NLP is an AI technology that understands unstructured human language and transforms voice or text communications into normalized data suitable for analysis.

Supply chain organizations are getting smarter in using internal data to optimize their networks, improve customer satisfaction and reduce costs. However, with the onset of COVID-19 in early 2020, it is clear that most supply chain organizations are not yet resilient enough.

Supply chains came across unprecedented challenges such as shortages of raw materials, unplanned operational shutdowns, labor shortages, delays in transportation, and a sudden rise or fall in demand. This caused major supply chain disruptions, which had a cascading effect. Products such as toilet paper, bread makers and home exercise equipment saw a sudden spike in demand, in some cases by as much as 600%. This resulted in temporary shortages for these products. At the same time, the demand for gasoline plunged due to a mix of geopolitical factors and COVID‑19‑related stay‑at‑home orders enacted around the world. The shock to oil and gas supply chains was both unforeseen and immediate.

(Chapter breaker)
1

Chapter 1

Why existing supply chain strategies fall short during a crisis

Amid unprecedented challenges, pivot to digital capabilities to understand external events.

Since most supply chain organizations use internal data to track their demand‑supply balances, they were not able to foresee the external events that impacted supply chains. The challenges related to COVID‑19 forced supply chains to react to the situation instead of being proactive. Most organizations took too long to recover from this shock, which impacted their operations and bottom line.

Supply chain organizations must look beyond their current landscape and use digital capabilities to build resiliency to handle challenges around crises like the COVID‑19 pandemic. Natural language processing is one of the emerging artificial intelligence applications that can help build supply chain resiliency.

(Chapter breaker)
2

Chapter 2

Understanding natural language processing

Use web scraping and social media listening to analyze data and gain business intelligence.

Natural language processing is a type of artificial intelligence technology, and it’s a component of text analytics. It focuses on understanding unstructured human language and transforming voice or text communications into normalized, structured data suitable for analysis.

NLP has a wide range of applications, and some familiar examples from our daily lives are online shopping product recommendations, voice-activated virtual assistants, and voice-to-text messaging.  

Siri and Alexa recognize and interpret human voice commands. Similarly, voice‑to‑text messaging interprets the human voice and converts it to text. Amazon, on the other hand, uses algorithms to “scrape” a user’s purchase history, along with items the user has liked and rated to make product recommendations. These are all applications of NLP.

Web scraping means extracting web content to track events and data that can have potential impact on business operations.

One example of web scraping is tracking the news for hurricanes, tornadoes and storms to assess the impact of these weather events on business operations. This helps businesses to take proactive measures before the actual event.

Other applications such as social media listening can provide valuable insights on analyzing data to gain business intelligence.

Social media listening means tracking mentions and conversations on social media platforms such as Twitter and Facebook. Software is used to monitor certain key words or phrases in social media posts, and an algorithm is used to analyze the data.

One example of social media listening is when marketing companies track social media conversations regarding their brand. This helps them to understand why, where and how these conversations are happening, and what people think about their brand.

Social media listening and web scraping can provide valuable insights that help businesses to be proactive and adjust their strategy.

Social media listening and web scraping can provide valuable insights that help businesses be proactive and adjust their strategy.

(Chapter breaker)
3

Chapter 3

How supply chain organizations can benefit from the use of NLP

Drive process efficiency to handle crisis situations such as the COVID-19 pandemic.

Supply chain organizations can gain many benefits from the use of NLP, including the following use cases:

1. Automate manual processes

NLP can be used to read thousands of shipment documents to provide valuable summary information for making targeted logistics improvements.

2. Monitor data changes

NLP algorithms can track real‑time changes in internal data, which can help in accurately maintaining master data.

3. Gather industry benchmarks

Web scraping can be used to gather industry benchmark data for transportation rates, fuel prices and labor costs. This data helps organizations to benchmark their performance against industry standards and identify cost‑saving opportunities.

4. Reduce language barriers

Most organizations have global operations, and language barriers can hinder process efficiency. NLP can be used to translate documents from one language to another to reduce regional language barriers.

5. Conduct compliance tracking

NLP can help verify compliance with ethical practices by web scraping publicly available information for potential breaches by suppliers.

In addition to the above benefits, supply chain organizations can use NLP for building resiliency to handle crisis situations such as the COVID-19 outbreak. The benefits of NLP can be realized in end‑to‑end supply chain processes, from supply side to logistics side.

Web scraping can be used to track critical external information related to key suppliers. Tracking a supplier’s stock market performance and assessing its financial statements can provide data on the supplier’s financial stability. Furthermore, web scraping and social media listening can provide other useful supplier data points in areas such as labor relations, regional governmental restrictions, local news regarding strikes or riots, and weather events that can impact operations of suppliers and may result in supply risk. Further analytics can be built on this data to create early warning indicators.

With these valuable insights, supply chain organizations can take proactive measures to rely more on alternate suppliers for risk mitigation. This can be helpful in crises such as the COVID‑19 pandemic, when some suppliers went out of business while others had to shut down operations due to governmental restrictions.

In addition to assessing supply risk, NLP also can be used to evaluate the impact of external events on manufacturing operations and logistics. During the COVID‑19 crisis, certain states noticed a spike in the number of infections, leading them to take measures such as closure of retail establishments and stay‑at‑home orders, which had a cascading effect on other supply chains. The COVID‑19 pandemic also impacted logistics operations due to a shortage of trucks for shipments and shipping port congestion, resulting in delayed shipments.

NLP can be used in these situations to track news about COVID‑19 cases and local government restrictions where an organization has operations. These data points can provide insights into potential future disruptions to operations. Similarly, NLP can be used to scrape data from logistics carriers’ websites and shipping port websites to assess the impact of a crisis and inform measures such as increasing safety stock inventory levels, using alternate modes of shipment and finding different routes to ship products.

NLP enables supply chain organizations to gather and monitor external data that can cause potential disruptions. Supply chains can mitigate risk by taking preventive measures and build resiliency to handle crises.

  • We deployed NLP at a Fortune 50 global manufacturing client to help protect its supply chain during the COVID-19 crisis.

    The project team primarily used NLP for internet news, social media and document searches to identify and interpret content that could have an impact on the client’s business. We applied this technique to understand what journalists are reporting and what users are saying at the web‑scale level.

    For social media content, we focused on analyzing tweets from Twitter. We identified and categorized tweets using an application programming interface that allowed us to see what vendors were officially saying, their employees’ sentiments, updates regarding the COVID‑19 pandemic, news on relevant material production in the local facilities, regional government restrictions and the general sentiment of the community in the client’s geographic area. This equipped our client to have more targeted conversations with its vendors, with a fuller understanding of current business conditions, and to plan accordingly.

    The EY team helped position the client for future opportunities in this rapidly evolving field, providing the following:

    • Ability to have targeted conversations with suppliers to provide continuity of supply
    • Insight-driven data to build further analytics to forecast demand and supply
    • Review of real‑time data for shipping port congestion to make necessary changes to priority shipments

Lead through the COVID-19 crisis

We have a clear view of the critical questions and new answers required for effective business continuity and resilience.

Explore

Contact us for immediate support

Gain access to our help with crisis management, business continuity and enterprise resilience.

 

Contact

Summary

Since most organizations use internal data to track their demand-supply balances, it’s challenging for them to foresee the external events that impact their supply chains. Enter natural language processing (NLP), an AI technology that transforms unstructured human language from voice or text into normalized data suitable for analysis. NLP can help companies monitor social media, news sources, and other webpages for external data on disruptive events as they begin to unfold, allowing them to mount a response to brace their supply chains sooner, rather than later.

About this article

By Sven Dharmani

EY Global Advanced Manufacturing & Mobility Supply Chain Leader

Passionate about transforming supply chains. Problem solver. Curious and collaborative. Avid traveler, scuba diver and car enthusiast.