It was a beautiful, sunny afternoon, and we were headed to one of our favorite weekend getaway destinations. My plan to get there was pretty straightforward; I knew the best route and all the usual traffic patterns, so we set out on what should have been a two-hour drive. Four hours later, we finally arrived, tired and frustrated — not a great way to kick off a relaxing weekend. Reflecting on where my planning went wrong, I relied almost exclusively on my personal knowledge of the route and traffic patterns – only internal and historical data. Of course, I took in some external data, like the weather conditions. However, I was still left exposed to risks from external disruptions like extra traffic from a major concert, road construction, and multiple car accidents clogging up the highway.
You can drive a car by looking in the rearview mirror
as long as nothing is ahead of you. – Bill Joy
Supply chain planning is much the same way. There is a tendency to rely heavily on historical trend data to find seasonal and buying patterns to help predict future results. Don’t get me wrong, historical data is vital to demand forecasting. However, the vast majority of volatility and uncertainty comes from the outside – from the extended network of supply, logistics, channel, and global trade partners. If this data is not included in your near-term planning calculations, it doesn’t matter what shiny new AI tool you use; the old adage still applies – garbage in, garbage out.
Open the data floodgates
So, the most effective way to create a comprehensive, accurate, and feasible supply chain plan is by creating a digital twin. A digital twin requires a tremendous amount of data from various internal and external sources to feed the AI-enabled insights engine.
For starters, you’ll want to consider data from all internal systems and data sources, including enterprise resource management systems (ERP), customer relationship management systems (CRM), manufacturing execution systems (MES), warehouse management (WMS), transportation management (TMS), product lifecycle management (PLM), finance and accounting, other planning systems, and more.
Then add external partner data across all tiers – not just the first tier (POS, store inventory, distribution center inventory, materials inventory, supplier capacity, and upsides, supply commits/decommits, advanced shipping notices), third-party data (weather, social sentiment, risk events, import and export duties and tariffs, restricted party lists, forced labor regulations, market data, and so on), and IoT sensor data (manufacturing lines, vehicles, storage units, etc.).
By now, you probably think this sounds theoretically sound, but there is not enough time and budget to make all those connections. Not to mention how you even begin to cleanse and harmonize data from many disparate sources.
Making multi-enterprise connections at speed and scale
Similar to how LinkedIn allows you to connect and maintain relationships with hundreds or thousands of peers, colleagues, and associates, a supply chain business network streamlines and scales the process of connecting to your ecosystem partners. A network has pre-established connections to hundreds of thousands of supply, channel, logistics, and global trade partners, speeding the time-to-value.
Take caution in evaluating supply chain business networks; they are not all created equal. They differ in terms of scope, scale, and capability. Look for supply chain business networks with the scope of all four ecosystems from complete visibility – supply, channel, logistics, and global trade. Another consideration is the scale of the network in terms of the number of connected enterprises and the depth of tiers. Some may only connect to the first or second tier. In contrast, effective networks can connect across all the tiers of upstream and downstream partners.
And lastly, best-in-class supply chain networks offer a full array of connection and integration capabilities. While you are advancing toward a total digital transformation, your partners may not have the resources to keep up. Your network provider should meet partners at their technical maturity rather than force a costly integration approach upon them. Of course, this includes the latest in Rest APIs and EDI, and it goes on to include a web portal and email-based integration for long-tail suppliers and partners.
Making dirty, disparate data into a valuable asset
While the network will shorten the time to establish connections and gather data to improve your planning process, it’s not enough. Data is inherently dirty, containing missing, wrong, and null values. And very often, different master data management (MDM) policies and practices are employed, even across internal systems. Bringing in data from outside the enterprise makes this challenge exponentially more difficult.
The multi-enterprise supply chain network needs a multi-enterprise MDM (ME-MDM) to transform disparate data into a valuable, decision-grade asset. Think of it as a universal translator that understands and interprets information from every system and partner into a single, clear language. To achieve ME-MDM, look for networks that sit on a platform with a canonical data model to manage the contextual attributes for each resource from all sources, including internal and ecosystem partner systems. The system must cleanse errant values and harmonize data from disparate multi-enterprise sources at scale.
—
Just as Bill Joy said, “You can drive a car by looking in the rearview mirror as long as nothing is ahead of you;” you can plan your supply chain using only historical data as long as there are no changes to demand or supply.
Mike Hitmar, Sr. Director, Product Marketing at e2open. Read more about how e2open can help you make feasible, profitable plans and increase agility with Connected Planning.
Effective Supply Chain Planning Is All In The Data
It was a beautiful, sunny afternoon, and we were headed to one of our favorite weekend getaway destinations. My plan to get there was pretty straightforward; I knew the best route and all the usual traffic patterns, so we set out on what should have been a two-hour drive. Four hours later, we finally arrived, tired and frustrated — not a great way to kick off a relaxing weekend. Reflecting on where my planning went wrong, I relied almost exclusively on my personal knowledge of the route and traffic patterns – only internal and historical data. Of course, I took in some external data, like the weather conditions. However, I was still left exposed to risks from external disruptions like extra traffic from a major concert, road construction, and multiple car accidents clogging up the highway.
Supply chain planning is much the same way. There is a tendency to rely heavily on historical trend data to find seasonal and buying patterns to help predict future results. Don’t get me wrong, historical data is vital to demand forecasting. However, the vast majority of volatility and uncertainty comes from the outside – from the extended network of supply, logistics, channel, and global trade partners. If this data is not included in your near-term planning calculations, it doesn’t matter what shiny new AI tool you use; the old adage still applies – garbage in, garbage out.
Open the data floodgates
So, the most effective way to create a comprehensive, accurate, and feasible supply chain plan is by creating a digital twin. A digital twin requires a tremendous amount of data from various internal and external sources to feed the AI-enabled insights engine.
For starters, you’ll want to consider data from all internal systems and data sources, including enterprise resource management systems (ERP), customer relationship management systems (CRM), manufacturing execution systems (MES), warehouse management (WMS), transportation management (TMS), product lifecycle management (PLM), finance and accounting, other planning systems, and more.
Then add external partner data across all tiers – not just the first tier (POS, store inventory, distribution center inventory, materials inventory, supplier capacity, and upsides, supply commits/decommits, advanced shipping notices), third-party data (weather, social sentiment, risk events, import and export duties and tariffs, restricted party lists, forced labor regulations, market data, and so on), and IoT sensor data (manufacturing lines, vehicles, storage units, etc.).
By now, you probably think this sounds theoretically sound, but there is not enough time and budget to make all those connections. Not to mention how you even begin to cleanse and harmonize data from many disparate sources.
Making multi-enterprise connections at speed and scale
Similar to how LinkedIn allows you to connect and maintain relationships with hundreds or thousands of peers, colleagues, and associates, a supply chain business network streamlines and scales the process of connecting to your ecosystem partners. A network has pre-established connections to hundreds of thousands of supply, channel, logistics, and global trade partners, speeding the time-to-value.
Take caution in evaluating supply chain business networks; they are not all created equal. They differ in terms of scope, scale, and capability. Look for supply chain business networks with the scope of all four ecosystems from complete visibility – supply, channel, logistics, and global trade. Another consideration is the scale of the network in terms of the number of connected enterprises and the depth of tiers. Some may only connect to the first or second tier. In contrast, effective networks can connect across all the tiers of upstream and downstream partners.
And lastly, best-in-class supply chain networks offer a full array of connection and integration capabilities. While you are advancing toward a total digital transformation, your partners may not have the resources to keep up. Your network provider should meet partners at their technical maturity rather than force a costly integration approach upon them. Of course, this includes the latest in Rest APIs and EDI, and it goes on to include a web portal and email-based integration for long-tail suppliers and partners.
Making dirty, disparate data into a valuable asset
While the network will shorten the time to establish connections and gather data to improve your planning process, it’s not enough. Data is inherently dirty, containing missing, wrong, and null values. And very often, different master data management (MDM) policies and practices are employed, even across internal systems. Bringing in data from outside the enterprise makes this challenge exponentially more difficult.
The multi-enterprise supply chain network needs a multi-enterprise MDM (ME-MDM) to transform disparate data into a valuable, decision-grade asset. Think of it as a universal translator that understands and interprets information from every system and partner into a single, clear language. To achieve ME-MDM, look for networks that sit on a platform with a canonical data model to manage the contextual attributes for each resource from all sources, including internal and ecosystem partner systems. The system must cleanse errant values and harmonize data from disparate multi-enterprise sources at scale.
—
Just as Bill Joy said, “You can drive a car by looking in the rearview mirror as long as nothing is ahead of you;” you can plan your supply chain using only historical data as long as there are no changes to demand or supply.
Mike Hitmar, Sr. Director, Product Marketing at e2open. Read more about how e2open can help you make feasible, profitable plans and increase agility with Connected Planning.
TAGS
Subscribe to Our YouTube Channel
TOPICS
Categories
Subscribe to Our Podcast
TRENDING POSTS
Sponsors