In the face of uncertain and ever-changing market dynamics, supply chain planning (SCP) must begin to shift from manual to data-driven automated processes to better support organizations’ need for agility and resiliency via improvements in both efficiency and accuracy.
According to respondents in a Gartner survey, 44% of supply chain professionals who have undertaken new supply chain planning implementation or improvement projects report an improvement in plan accuracy. Thirty percent also report an improvement in speed of decision making, with the average improvement reported for speed of decision making being 13%.
The emergence of generative AI solutions like ChatGPT and availability of machine learning capabilities has captivated many organizations. As companies consider investment, supply chain leaders would be well-served to focus their attention now on preparing their data to maximize potential benefit.
Evaluate and Categorize Data Needs
To effectively deploy emerging data-dependent solutions like GenAI, leaders must first evaluate their current supply chain data estate, and then develop a method for extending their planning capabilities further by evaluating incremental data sources. They should catalog their current and expected data needs across three categories.
First-party data: Created and used across and within the organization, it’s typically from sourcing and procurement, sales and marketing and manufacturing and distribution. First-party data represents the commonly used and vital information set required, at a bare minimum, to roughly deliver industry average supply chain performance based on valid internal models.
Second-party data: External channel partner decisions and processes produce this data. To drive performance forward, supply chain planning organizations increasingly need to leverage collaborative planning functions and data-sharing from suppliers and customers. This will deliver a more accurate digital representation of the physical supply chain.
Third-party data: This is data which is sourced from external entities other than direct customers and suppliers. It can include commercial data, such as syndicated data including demographic or customer focus group analyses and publicly available data, such as macroeconomic indicators. This data can improve forecast accuracy and provide improved supply chain planning decision support.
Prioritize Use Cases and Define Processes
Once leaders understand their data needs, they should evaluate the quality of the data and types of challenges that need to be solved to move overall supply chain performance forward. Supply chain leaders typically have a set of priority issues to be addressed, such as subpar service levels, excess inventory or inefficient execution coordination.
In these cases, leaders should identify what kind of information would deliver the desired process improvements. Consideration should also be given to setting a cadence for making data available, from which sources and for whom in the organization. As leaders move through these steps, they may realize that their existing data estate contains significant issues related to availability, quality and latency.
These foundational issues can often be addressed via improved middleware/data messaging services, as well as the implementation of a data governance group. By addressing these foundational issues, leaders will be best situated to leverage incremental data sources to move their supply chain performance forward as they seek to move past Level 3 process maturity.
Incorporate Additional Data-Centric Strategies
Once the foundational data estate is addressed, leaders can begin to assess how best to incorporate more data into their planning process. As a starting point, consider opportunities for inclusion of second-party data, typically beginning with customer collaboration, followed by supplier collaboration. These capabilities can be further extended, for example, to include development of demand-sensing solutions to drive further supply chain performance improvements.
From there, they can look to incorporate third-party data, such as transportation visibility solutions or other supply chain risk analytics data sources. These data sources will become critical as the organization seeks to build towards a true digital supply chain twin.
By establishing the correct foundational infrastructure components and gaining an understanding of their data needs, leaders will be best poised to effectively deploy technology into the existing supply chain system architecture, and to ensure adequate alignment of current and future process requirements.