Coding Knowledge Can Aid Supply Chain Decisions — and Opportunities

February 13, 2023
By Fabiola Marinkov, Drew Childs

In 2018, Forbes estimated that around 2.5 quintillion bytes of data were created every day. Fast forward to today and imagine how much more is being collected, especially with supply chains and Internet of Things (IoT) devices generating raw data points across every phase of business.

All that information is collected and stored to be analyzed and help improve processes. With the proper data points, a business can reduce production time, provide procurement cost savings and cut logistics lead times, among other benefits, leading to an organization that maintains relevance in its industry.

However, raw data alone won’t provide the insights necessary for real changes to occur. Employees with the tools to wrangle, clean, and visualize millions of bytes of data is becoming a pivotal skill within supply chains.

Programming Languages

After it’s collected, data generally needs to be extracted, transformed, cleaned, loaded to a database and then visualized. Throughout this process, programming languages and coding play a significant role.

Today, there are more than 250 programming languages, but not all are geared toward data analytics. The three most significant ones for the field of data science are:

Python. According to an O’Reilly survey, more than 40 percent of data scientists claim they use Python as their main coding language. It is a strategic, scalable and dynamic language that supports multiple paradigms from functional to structured and procedural programming. Python has a great variety of open-source data science libraries like NumPy, Pandas and Scikit-learn, making data manipulation that much easier.

The large community of Python programmers constantly contribute to the language and make the learning curve quicker for newbies. Overall, the resources and functionality of this language make it ideal for data analysis.

Scala. A flexible and concise, general-purpose programming language, Scala was originally developed to address issues with Java. Combining object-oriented and functional programming languages, it was built to implement scalable solutions to produce actionable insights on big data.

Often used on huge data sets, especially when data engineering is involved, Scala is an effective language that allows parallel processing and has additional applications for machine learning and web programming. Scala coding for data science was created to describe common programming patterns in an expressive and type-safe manner.

SQL. Structured query language (SQL) is a popular domain-specific language, perfect for managing data. It is not exclusively used for data science but is quite useful when handling data management systems and “calling” data with specific queries. It is domain-specific and perfect for incorporating relations among entities and variables.

The scope of SQL includes querying, data definition, manipulation and access control, making it a convenient solution for storing and retrieving data in relational databases. It can also be a great base for solutions created in business intelligence tools. In supply-chain related professions, SQL makes a perfect base knowledge regardless of what additional programming aspiration one might have.

Supply Chain Data Professions

With an abundance of data available across the supply chain, professionals at all levels can benefit from incorporating coding and advanced analytics into their skill sets, including:

Operations managers. If the processes within the organization are not operating optimally, it is often very difficult for the business to keep up with competitors. That is why operations managers need to evaluate data points collected from these processes and find ways to increase productivity or improve the product offering.

Scala is great for managing large amounts of data and providing quick data retrieval. In this case, the language can be implemented to rapidly gather all the data points generated from operations and use them for real-time decision-making. Data-driven decisions within operations keep businesses flowing and relevant in the market.

Procurement leads. Sourcing materials or services for a business’ daily operations includes negotiating contracts, purchasing items or services, and keeping a record of all transactions. With that much involvement in a company’s spend, there is a huge opportunity for analyzing that spend to improve purchasing decisions.

SQL can be used to pull specific accounts payable/purchase order data from the database for additional analysis. Procurement leads can then use that data to better analyze client relationships and improve payment terms. This information can also be used to analyze how much spend is under contract and look to implement a catalog for products and services that need to be managed. Procurement analytics overall can improve spend behavior within a business and lead to cost savings.

Logistics analysts, who constantly evaluate lead times and can use data to identify common factors that impact deliveries. Data can also be collected and used for warehousing key products, being sure to have the necessary inventory on hand while also minimizing holding costs.

For such projects, Python can be extremely useful in cleaning the data and creating visuals. The visualizations can then be analyzed and used for optimizing those processes and driving business performance. Data analysis may not be the full workload of logistics professionals, but it can have a huge impact on improving the areas they manage.

The use of coding for analytics can open opportunities in any job role along the supply chain. Data is everywhere, and Python, Scala and SQL coding languages are needed everywhere. Learning any one of these languages will take time and effort, but the reward of solid, data-driven decisions far exceeds the cost.

(Image credit: Getty Images/Yuichiro Chino)

Fabiola Marinkov

Fabiola Marinkov is delivery lead at IBM.

Drew Childs

Drew Childs is a member of IBM’s Procurement Analytics as a Service team. The perspective and opinions represented are those of the author(s) and do not represent those of IBM; they reflect experiences at various companies and organizations.