Data Freshness: Your Hack to Real Time Demand & Supply Planning
Data Freshness: Your Hack to Real Time Demand & Supply Planning
Introduction:
Supply chains today span across nations making end to end visibility both a necessity and a challenge. Be it a geopolitical disruption or a marketing promotion, they all have a ripple effect on your supply chain. But how does one quantify this impact and act upon it?
According to McKinsey, “Applying AI-driven forecasting to supply chain management, can reduce errors by between 20 and 50 percent—and translate into a reduction in lost sales and product unavailability of up to 65 percent.”
Well, thanks to AI powered supply chain solutions that make real time demand and supply planning a reality. With multiple layers of both internal and external data these solutions sense demand signals continuously and plan supply in real time. They equip businesses with better visibility into their supply chain which allows them to proactively prepare for any future disruptions. Having said that, the success of any data driven supply chain lies in the freshness of data collected and the way it is used to enable quick decision making.
.png)
Why is data freshness important?
A traditional demand forecasting that is done at the beginning of the month would no longer be relevant as you are halfway through the month. This may lead to an out-of-stock or an inventory pileup situation. As a result, you would end up losing sales revenue or stagnating working capital. This is why businesses have to account for important real time information such as changes in price, marketing/promotional events, change in weather and holidays in addition to historical sales data. It helps achieve better demand forecasting accuracy which eventually helps plan supply effectively.
Now, let’s understand what goes into implementing a real time supply chain planning system.
Demand Sensing
Demand sensing is an enhancement of the conventional demand forecasting process which provides more accurate and real time predictions. This allows businesses to proactively identify any outliers and cope with them before they turn into a threat. Kronoscope enables demand sensing by taking into account the following data signals:
Historical Data & Demand Patterns
Kronoscope determines baseline demand predictions by factoring in data points like historical trends, seasonal effects, cyclicity and outlier corrections. It analyses historical demand data to identify any growth or decline trends for every SKU in a particular geography. It also studies long term seasonal effects by establishing a relationship between sales data and different seasons like summer, winter, fall and spring. This helps identify certain SKUs that have seasonal peaks and forecast their demand levels with ease. Similarly, Kronoscope identifies short term cyclicity of different SKUs based on their sales. This could include the first day of the week effect, weekdays vs weekends, end of the month effect on the sales of each SKU. Further, it looks for any abnormalities/outliers in the historical data and neglects that in order to offer more accurate demand predictions.
Pricing & Promotions
Changes in pricing and any promotional activities are included to enhance the baseline numbers and arrive at the Factor adjusted demand predictions. Kronoscope allows users to run price based simulations on the system generated demand forecasts. The users can see how demand changes to a change in price. This is done by establishing a price elasticity curve that captures the price at which a product was sold at different points in time. It assists businesses to determine and recommend price points that will help them meet revenue targets.
Kronoscope allows users to capture any promotional events including the event type, impact segment and the duration. This is then used to accurately predict demand for similar/recurring promotional events in the future. Kronoscope’s AI engine analyzes historical trends and the events calendar to identify the best possible demand patterns and recommend a promotional uplift. These enhanced forecasts then flow into inventory planning.
Similar SKU/Store Attributes
Kronoscope also offers similarity adjusted demand predictions for newly launched products or even stores. This is done by intelligently associating similar/existing product attributes. For newly launched stores, the SKU level demands at similar/existing stores are associated to predict demand accurately.
Real time Inventory Planning
One of the most valuable features of an automated supply planning system is the continuous adjustment of inventory and safety norms based on real time demand predictions. Businesses would be able to implement these changes across thousands of SKUs within multiple stores in no time. This helps in achieving service level targets consistently and boost revenue irrespective of the dynamic demand levels. Kronoscope automates purchase and replenishment planning by taking into account these major data points:
Dynamic Safety Stock Adjustments
Maintaining adequate safety stock levels is crucial for any inventory driven business. Every business desires to have an uninterrupted flow of operations to ensure product availability at all times. Nevertheless, there are situations when your products run out of stock unexpectedly. This is when safety stock comes in handy to meet the excess demand and not lose the revenue opportunity. It is also important to maintain the right safety stock levels for each SKU and adjust them dynamically as inventory levels, shelf life, lead times and fill rates fluctuate. Kronoscope recommends safety stock levels dynamically based on the combination of two major factors - Demand Variability and Service Level Targets.
Demand variability
Demand variability is calculated by comparing the difference between the actual and the predicted demand levels for any particular SKU. SKUs with higher demand variability would require a higher safety stock level and vice versa.
Service level targets
Service level targets are used to classify SKUs into ABC categories. Category A products are the ones that yield maximum revenue and should have service targets around 95%. Category B products have service levels between 90-80%. And category C has the long tail products that have service levels between 80-75%. This will ensure consistent availability for the fast moving products and prevent pile up of the long tail products.
Changes in Supplier Metrics
Kronoscope helps determine purchase order quantities by factoring in dynamically changing demand predictions, current inventory, supplier metrics, safety stock, open orders and inter store transfers for each SKUs under different stores/warehouses. When there is a shortage in supply due to prolonged lead time or reduced fill rates, the system recommends a higher safety stock in order to make up for the delay in delivery and back orders. The purchase order quantity for that period is also adjusted accordingly. Once the orders are delivered, all inventory and safety norms are readjusted according to the updated inventory levels.
Further, Kronoscope also enables real time supplier selection across SKUs. All suppliers for a specific SKU are scored based on lead time, fill rate and price. This allows the user to choose the ideal supplier as per their dynamic business requirements.
Inventory Ageing
When dealing with perishable products, it is crucial to account for inventory aging in order to avoid wastage. Kronoscope uses the FIFO approach to keep track of inventory’s expiry date and suggest which one has to be moved out first. This simplifies movement of inventory even when there are multiple/short inventory turns across SKUs.
Inventory Rebalancing
Kronoscope with its inventory control tower feature gives out potential stock outs and potential inventory pile up alerts. Based on these inventory norms, it will also suggest transfer orders between different SKUs in order to maintain optimal inventory levels.
In conclusion, ensuring data freshness boosts the capability of a supply planning system to predict demand levels accurately and optimize inventory in real time. These capabilities are further enhanced when artificial intelligence and human intuition are integrated to make supply chain decisions.
Data Freshness: Your Hack to Real Time Demand & Supply Planning
Introduction:
Supply chains today span across nations making end to end visibility both a necessity and a challenge. Be it a geopolitical disruption or a marketing promotion, they all have a ripple effect on your supply chain. But how does one quantify this impact and act upon it?
According to McKinsey, “Applying AI-driven forecasting to supply chain management, can reduce errors by between 20 and 50 percent—and translate into a reduction in lost sales and product unavailability of up to 65 percent.”
Well, thanks to AI powered supply chain solutions that make real time demand and supply planning a reality. With multiple layers of both internal and external data these solutions sense demand signals continuously and plan supply in real time. They equip businesses with better visibility into their supply chain which allows them to proactively prepare for any future disruptions. Having said that, the success of any data driven supply chain lies in the freshness of data collected and the way it is used to enable quick decision making.
Why is data freshness important?
A traditional demand forecasting that is done at the beginning of the month would no longer be relevant as you are halfway through the month. This may lead to an out-of-stock or an inventory pileup situation. As a result, you would end up losing sales revenue or stagnating working capital. This is why businesses have to account for important real time information such as changes in price, marketing/promotional events, change in weather and holidays in addition to historical sales data. It helps achieve better demand forecasting accuracy which eventually helps plan supply effectively.
Now, let’s understand what goes into implementing a real time supply chain planning system.
Demand Sensing
Demand sensing is an enhancement of the conventional demand forecasting process which provides more accurate and real time predictions. This allows businesses to proactively identify any outliers and cope with them before they turn into a threat. Kronoscope enables demand sensing by taking into account the following data signals:
Historical Data & Demand Patterns
Kronoscope determines baseline demand predictions by factoring in data points like historical trends, seasonal effects, cyclicity and outlier corrections. It analyses historical demand data to identify any growth or decline trends for every SKU in a particular geography. It also studies long term seasonal effects by establishing a relationship between sales data and different seasons like summer, winter, fall and spring. This helps identify certain SKUs that have seasonal peaks and forecast their demand levels with ease. Similarly, Kronoscope identifies short term cyclicity of different SKUs based on their sales. This could include the first day of the week effect, weekdays vs weekends, end of the month effect on the sales of each SKU. Further, it looks for any abnormalities/outliers in the historical data and neglects that in order to offer more accurate demand predictions.
Pricing & Promotions
Changes in pricing and any promotional activities are included to enhance the baseline numbers and arrive at the Factor adjusted demand predictions. Kronoscope allows users to run price based simulations on the system generated demand forecasts. The users can see how demand changes to a change in price. This is done by establishing a price elasticity curve that captures the price at which a product was sold at different points in time. It assists businesses to determine and recommend price points that will help them meet revenue targets.
Kronoscope allows users to capture any promotional events including the event type, impact segment and the duration. This is then used to accurately predict demand for similar/recurring promotional events in the future. Kronoscope’s AI engine analyzes historical trends and the events calendar to identify the best possible demand patterns and recommend a promotional uplift. These enhanced forecasts then flow into inventory planning.
Similar SKU/Store Attributes
Kronoscope also offers similarity adjusted demand predictions for newly launched products or even stores. This is done by intelligently associating similar/existing product attributes. For newly launched stores, the SKU level demands at similar/existing stores are associated to predict demand accurately.
Real time Inventory Planning
One of the most valuable features of an automated supply planning system is the continuous adjustment of inventory and safety norms based on real time demand predictions. Businesses would be able to implement these changes across thousands of SKUs within multiple stores in no time. This helps in achieving service level targets consistently and boost revenue irrespective of the dynamic demand levels. Kronoscope automates purchase and replenishment planning by taking into account these major data points:
Dynamic Safety Stock Adjustments
Maintaining adequate safety stock levels is crucial for any inventory driven business. Every business desires to have an uninterrupted flow of operations to ensure product availability at all times. Nevertheless, there are situations when your products run out of stock unexpectedly. This is when safety stock comes in handy to meet the excess demand and not lose the revenue opportunity. It is also important to maintain the right safety stock levels for each SKU and adjust them dynamically as inventory levels, shelf life, lead times and fill rates fluctuate. Kronoscope recommends safety stock levels dynamically based on the combination of two major factors - Demand Variability and Service Level Targets.
Demand variability
Demand variability is calculated by comparing the difference between the actual and the predicted demand levels for any particular SKU. SKUs with higher demand variability would require a higher safety stock level and vice versa.
Service level targets
Service level targets are used to classify SKUs into ABC categories. Category A products are the ones that yield maximum revenue and should have service targets around 95%. Category B products have service levels between 90-80%. And category C has the long tail products that have service levels between 80-75%. This will ensure consistent availability for the fast moving products and prevent pile up of the long tail products.
Changes in Supplier Metrics
Kronoscope helps determine purchase order quantities by factoring in dynamically changing demand predictions, current inventory, supplier metrics, safety stock, open orders and inter store transfers for each SKUs under different stores/warehouses. When there is a shortage in supply due to prolonged lead time or reduced fill rates, the system recommends a higher safety stock in order to make up for the delay in delivery and back orders. The purchase order quantity for that period is also adjusted accordingly. Once the orders are delivered, all inventory and safety norms are readjusted according to the updated inventory levels.
Further, Kronoscope also enables real time supplier selection across SKUs. All suppliers for a specific SKU are scored based on lead time, fill rate and price. This allows the user to choose the ideal supplier as per their dynamic business requirements.
Inventory Ageing
When dealing with perishable products, it is crucial to account for inventory aging in order to avoid wastage. Kronoscope uses the FIFO approach to keep track of inventory’s expiry date and suggest which one has to be moved out first. This simplifies movement of inventory even when there are multiple/short inventory turns across SKUs.
Inventory Rebalancing
Kronoscope with its inventory control tower feature gives out potential stock outs and potential inventory pile up alerts. Based on these inventory norms, it will also suggest transfer orders between different SKUs in order to maintain optimal inventory levels.
In conclusion, ensuring data freshness boosts the capability of a supply planning system to predict demand levels accurately and optimize inventory in real time. These capabilities are further enhanced when artificial intelligence and human intuition are integrated to make supply chain decisions.
Data Freshness: Your Hack to Real Time Demand & Supply Planning
Introduction:
Supply chains today span across nations making end to end visibility both a necessity and a challenge. Be it a geopolitical disruption or a marketing promotion, they all have a ripple effect on your supply chain. But how does one quantify this impact and act upon it?
According to McKinsey, “Applying AI-driven forecasting to supply chain management, can reduce errors by between 20 and 50 percent—and translate into a reduction in lost sales and product unavailability of up to 65 percent.”
Well, thanks to AI powered supply chain solutions that make real time demand and supply planning a reality. With multiple layers of both internal and external data these solutions sense demand signals continuously and plan supply in real time. They equip businesses with better visibility into their supply chain which allows them to proactively prepare for any future disruptions. Having said that, the success of any data driven supply chain lies in the freshness of data collected and the way it is used to enable quick decision making.
Why is data freshness important?
A traditional demand forecasting that is done at the beginning of the month would no longer be relevant as you are halfway through the month. This may lead to an out-of-stock or an inventory pileup situation. As a result, you would end up losing sales revenue or stagnating working capital. This is why businesses have to account for important real time information such as changes in price, marketing/promotional events, change in weather and holidays in addition to historical sales data. It helps achieve better demand forecasting accuracy which eventually helps plan supply effectively.
Now, let’s understand what goes into implementing a real time supply chain planning system.
Demand Sensing
Demand sensing is an enhancement of the conventional demand forecasting process which provides more accurate and real time predictions. This allows businesses to proactively identify any outliers and cope with them before they turn into a threat. Kronoscope enables demand sensing by taking into account the following data signals:
Historical Data & Demand Patterns
Kronoscope determines baseline demand predictions by factoring in data points like historical trends, seasonal effects, cyclicity and outlier corrections. It analyses historical demand data to identify any growth or decline trends for every SKU in a particular geography. It also studies long term seasonal effects by establishing a relationship between sales data and different seasons like summer, winter, fall and spring. This helps identify certain SKUs that have seasonal peaks and forecast their demand levels with ease. Similarly, Kronoscope identifies short term cyclicity of different SKUs based on their sales. This could include the first day of the week effect, weekdays vs weekends, end of the month effect on the sales of each SKU. Further, it looks for any abnormalities/outliers in the historical data and neglects that in order to offer more accurate demand predictions.
Pricing & Promotions
Changes in pricing and any promotional activities are included to enhance the baseline numbers and arrive at the Factor adjusted demand predictions. Kronoscope allows users to run price based simulations on the system generated demand forecasts. The users can see how demand changes to a change in price. This is done by establishing a price elasticity curve that captures the price at which a product was sold at different points in time. It assists businesses to determine and recommend price points that will help them meet revenue targets.
Kronoscope allows users to capture any promotional events including the event type, impact segment and the duration. This is then used to accurately predict demand for similar/recurring promotional events in the future. Kronoscope’s AI engine analyzes historical trends and the events calendar to identify the best possible demand patterns and recommend a promotional uplift. These enhanced forecasts then flow into inventory planning.
Similar SKU/Store Attributes
Kronoscope also offers similarity adjusted demand predictions for newly launched products or even stores. This is done by intelligently associating similar/existing product attributes. For newly launched stores, the SKU level demands at similar/existing stores are associated to predict demand accurately.
Real time Inventory Planning
One of the most valuable features of an automated supply planning system is the continuous adjustment of inventory and safety norms based on real time demand predictions. Businesses would be able to implement these changes across thousands of SKUs within multiple stores in no time. This helps in achieving service level targets consistently and boost revenue irrespective of the dynamic demand levels. Kronoscope automates purchase and replenishment planning by taking into account these major data points:
Dynamic Safety Stock Adjustments
Maintaining adequate safety stock levels is crucial for any inventory driven business. Every business desires to have an uninterrupted flow of operations to ensure product availability at all times. Nevertheless, there are situations when your products run out of stock unexpectedly. This is when safety stock comes in handy to meet the excess demand and not lose the revenue opportunity. It is also important to maintain the right safety stock levels for each SKU and adjust them dynamically as inventory levels, shelf life, lead times and fill rates fluctuate. Kronoscope recommends safety stock levels dynamically based on the combination of two major factors - Demand Variability and Service Level Targets.
Demand variability
Demand variability is calculated by comparing the difference between the actual and the predicted demand levels for any particular SKU. SKUs with higher demand variability would require a higher safety stock level and vice versa.
Service level targets
Service level targets are used to classify SKUs into ABC categories. Category A products are the ones that yield maximum revenue and should have service targets around 95%. Category B products have service levels between 90-80%. And category C has the long tail products that have service levels between 80-75%. This will ensure consistent availability for the fast moving products and prevent pile up of the long tail products.
Changes in Supplier Metrics
Kronoscope helps determine purchase order quantities by factoring in dynamically changing demand predictions, current inventory, supplier metrics, safety stock, open orders and inter store transfers for each SKUs under different stores/warehouses. When there is a shortage in supply due to prolonged lead time or reduced fill rates, the system recommends a higher safety stock in order to make up for the delay in delivery and back orders. The purchase order quantity for that period is also adjusted accordingly. Once the orders are delivered, all inventory and safety norms are readjusted according to the updated inventory levels.
Further, Kronoscope also enables real time supplier selection across SKUs. All suppliers for a specific SKU are scored based on lead time, fill rate and price. This allows the user to choose the ideal supplier as per their dynamic business requirements.
Inventory Ageing
When dealing with perishable products, it is crucial to account for inventory aging in order to avoid wastage. Kronoscope uses the FIFO approach to keep track of inventory’s expiry date and suggest which one has to be moved out first. This simplifies movement of inventory even when there are multiple/short inventory turns across SKUs.
Inventory Rebalancing
Kronoscope with its inventory control tower feature gives out potential stock outs and potential inventory pile up alerts. Based on these inventory norms, it will also suggest transfer orders between different SKUs in order to maintain optimal inventory levels.
In conclusion, ensuring data freshness boosts the capability of a supply planning system to predict demand levels accurately and optimize inventory in real time. These capabilities are further enhanced when artificial intelligence and human intuition are integrated to make supply chain decisions.
Introduction:
Supply chains today span across nations making end to end visibility both a necessity and a challenge. Be it a geopolitical disruption or a marketing promotion, they all have a ripple effect on your supply chain. But how does one quantify this impact and act upon it?
According to McKinsey, “Applying AI-driven forecasting to supply chain management, can reduce errors by between 20 and 50 percent—and translate into a reduction in lost sales and product unavailability of up to 65 percent.”
Well, thanks to AI powered supply chain solutions that make real time demand and supply planning a reality. With multiple layers of both internal and external data these solutions sense demand signals continuously and plan supply in real time. They equip businesses with better visibility into their supply chain which allows them to proactively prepare for any future disruptions. Having said that, the success of any data driven supply chain lies in the freshness of data collected and the way it is used to enable quick decision making.
Why is data freshness important?
A traditional demand forecasting that is done at the beginning of the month would no longer be relevant as you are halfway through the month. This may lead to an out-of-stock or an inventory pileup situation. As a result, you would end up losing sales revenue or stagnating working capital. This is why businesses have to account for important real time information such as changes in price, marketing/promotional events, change in weather and holidays in addition to historical sales data. It helps achieve better demand forecasting accuracy which eventually helps plan supply effectively.
Now, let’s understand what goes into implementing a real time supply chain planning system.
Demand Sensing
Demand sensing is an enhancement of the conventional demand forecasting process which provides more accurate and real time predictions. This allows businesses to proactively identify any outliers and cope with them before they turn into a threat. Kronoscope enables demand sensing by taking into account the following data signals:
Historical Data & Demand Patterns
Kronoscope determines baseline demand predictions by factoring in data points like historical trends, seasonal effects, cyclicity and outlier corrections. It analyses historical demand data to identify any growth or decline trends for every SKU in a particular geography. It also studies long term seasonal effects by establishing a relationship between sales data and different seasons like summer, winter, fall and spring. This helps identify certain SKUs that have seasonal peaks and forecast their demand levels with ease. Similarly, Kronoscope identifies short term cyclicity of different SKUs based on their sales. This could include the first day of the week effect, weekdays vs weekends, end of the month effect on the sales of each SKU. Further, it looks for any abnormalities/outliers in the historical data and neglects that in order to offer more accurate demand predictions.
Pricing & Promotions
Changes in pricing and any promotional activities are included to enhance the baseline numbers and arrive at the Factor adjusted demand predictions. Kronoscope allows users to run price based simulations on the system generated demand forecasts. The users can see how demand changes to a change in price. This is done by establishing a price elasticity curve that captures the price at which a product was sold at different points in time. It assists businesses to determine and recommend price points that will help them meet revenue targets.
Kronoscope allows users to capture any promotional events including the event type, impact segment and the duration. This is then used to accurately predict demand for similar/recurring promotional events in the future. Kronoscope’s AI engine analyzes historical trends and the events calendar to identify the best possible demand patterns and recommend a promotional uplift. These enhanced forecasts then flow into inventory planning.
Similar SKU/Store Attributes
Kronoscope also offers similarity adjusted demand predictions for newly launched products or even stores. This is done by intelligently associating similar/existing product attributes. For newly launched stores, the SKU level demands at similar/existing stores are associated to predict demand accurately.
Real time Inventory Planning
One of the most valuable features of an automated supply planning system is the continuous adjustment of inventory and safety norms based on real time demand predictions. Businesses would be able to implement these changes across thousands of SKUs within multiple stores in no time. This helps in achieving service level targets consistently and boost revenue irrespective of the dynamic demand levels. Kronoscope automates purchase and replenishment planning by taking into account these major data points:
Dynamic Safety Stock Adjustments
Maintaining adequate safety stock levels is crucial for any inventory driven business. Every business desires to have an uninterrupted flow of operations to ensure product availability at all times. Nevertheless, there are situations when your products run out of stock unexpectedly. This is when safety stock comes in handy to meet the excess demand and not lose the revenue opportunity. It is also important to maintain the right safety stock levels for each SKU and adjust them dynamically as inventory levels, shelf life, lead times and fill rates fluctuate. Kronoscope recommends safety stock levels dynamically based on the combination of two major factors - Demand Variability and Service Level Targets.
Demand variability
Demand variability is calculated by comparing the difference between the actual and the predicted demand levels for any particular SKU. SKUs with higher demand variability would require a higher safety stock level and vice versa.
Service level targets
Service level targets are used to classify SKUs into ABC categories. Category A products are the ones that yield maximum revenue and should have service targets around 95%. Category B products have service levels between 90-80%. And category C has the long tail products that have service levels between 80-75%. This will ensure consistent availability for the fast moving products and prevent pile up of the long tail products.
Changes in Supplier Metrics
Kronoscope helps determine purchase order quantities by factoring in dynamically changing demand predictions, current inventory, supplier metrics, safety stock, open orders and inter store transfers for each SKUs under different stores/warehouses. When there is a shortage in supply due to prolonged lead time or reduced fill rates, the system recommends a higher safety stock in order to make up for the delay in delivery and back orders. The purchase order quantity for that period is also adjusted accordingly. Once the orders are delivered, all inventory and safety norms are readjusted according to the updated inventory levels.
Further, Kronoscope also enables real time supplier selection across SKUs. All suppliers for a specific SKU are scored based on lead time, fill rate and price. This allows the user to choose the ideal supplier as per their dynamic business requirements.
Inventory Ageing
When dealing with perishable products, it is crucial to account for inventory aging in order to avoid wastage. Kronoscope uses the FIFO approach to keep track of inventory’s expiry date and suggest which one has to be moved out first. This simplifies movement of inventory even when there are multiple/short inventory turns across SKUs.
Inventory Rebalancing
Kronoscope with its inventory control tower feature gives out potential stock outs and potential inventory pile up alerts. Based on these inventory norms, it will also suggest transfer orders between different SKUs in order to maintain optimal inventory levels.
In conclusion, ensuring data freshness boosts the capability of a supply planning system to predict demand levels accurately and optimize inventory in real time. These capabilities are further enhanced when artificial intelligence and human intuition are integrated to make supply chain decisions.
Like what you read?
Subscribe to receive a monthly digest of our most valuable resources like blog posts, whitepapers and much more
.png)
%20(1).png)
.png)