Article

What is ATP and how to get started

Use an Available-to-Promise system to make effective order fulfilment

Published

June 2019

Download the article

Author

Effective order fulfilment is the ability to combine planning and execution capabilities in the supply chain so that you ensure reliable order promising and improve customer service levels. You can support your effective order fulfilment with a tool called ATP, and together they can make the difference for achieving customer satisfaction.

What is order fulfilment?

Oracle and Capgemini have conducted a study with 589 supply chain executives showing that the number one challenge in maintaining customer satisfaction is “Accurately promising dates based on fulfilment planning lead times/estimates” (Peerless Research Group, 2016).

If you have a good order fulfilment process, it can make all the difference when it comes to achieving superior customer service, allocating orders to the right customers, delivering at the right time and optimising stock consumption.

One of the most important things to do well is promising customer orders. The ability to promise orders to customers validly – with dates and quantities that can be met can be a significant competitive weapon.

(Wallace & Stahl, 2003)

However, if your order fulfilment is imperfect, the cost will be decreased customer satisfaction and inefficient, suboptimal allocation of orders to customers (considering profitability). This will result in lost revenue as well as an increase in supply chain costs due to expediting to meet requested customer delivery dates, inventory issues and the number of resources required to keep up with the backlog in orders.

We all know that through superior customer service, companies can build strong and lasting relationships with their customers. This concept and the order fulfilment process itself seem simple at first sight: You receive an order and enter it into the system. Next, you send the order to the warehouse for pickup (in case of make-to-stock and to production for make-to-order), and the order is picked and packed for shipping, after which you ship the order to the customer. In this process, we assume that the product is already available in the warehouse (make-to-stock) or production facility, and raw materials are readily available for production (make-to-order). So, throughout the order fulfilment process, a company identifies what it has or will have available to sell, when it can be ready, and how it will get to the customer.

If it is so simple, then why do so many companies still struggle with delivering to their customers on time and in full?

Figure 1: A simple representation of the order fulfilment process

Order fulfilment starts before the order is received

Many companies still struggle with their order fulfilment process even though it seems like a simple process at first. The reason is that order fulfilment actually starts a long time before the customer order reaches the company.

There are several enabling processes that need to take place beforehand:

  • Planning demand in advance to ensure supply planning gets the right signals
  • Managing inventory and visibility of inventory levels to ensure customer service levels are up to par and orders can be promised based on realistic data
  • Supply planning to ensure products are produced in time and of high quality
  • Delivery planning (such as contracts and agreements with logistics providers) to ensure that the product reaches the customer site or pickup point

Or, said differently: If you want a good order fulfilment process, it all starts with tackling the imbalance between demand and supply, an imbalance that leads to increased backlogs, decreased customer service, higher inventory costs, costs for urgent order handling etc. This is clearly a situation that can be very dangerous for a business and a situation that needs to be anticipated well ahead of time.

Figure 2: A simple representation of the order fulfilment process

Balancing supply and demand as a predecessor for ensuring product availability

When you balance supply and demand, it happens in different processes and at different levels and dimensions. One of the processes that has an important impact on order fulfilment is called “master planning”. In this process, the planner ensures that the Sales and Operations Plan is translated into a realistic operational plan by constraining the Sales and Operations Plan and stabilising it towards production and procurement.

The main focus of this process is to create stability and product availability by balancing supply with demand, e.g. by levelling demand towards a stable production plan. It helps create a prioritisation between customers by ensuring supply through overtime or time off in production or use of temporary hires and by building up or reducing stock. As a result, you get the master production schedule, which is a plan that deals with what mix of products is going to be produced when, as it applies supply constraints (such as capacity and resources) to the forecasted demand plan. Based on this schedule, you can make production estimates and stock projections at a product mix level, which forms the basis of the order fulfilment process. The constrained plan gives you an indication of your future expected supply and the constraints you need to consider when promising orders.

Available-to-Promise systems can be considered as tools that can help in efficient decision-making related to order fulfilment processes.

 

Available to Promise

If you look at the traditional definition of the order fulfilment process, it starts with an order coming in and being registered in the system – so what happens next? To ensure that you make the right decisions, you need to identify your stock levels and incoming production, analyse incoming orders and prioritise orders if needed. Many companies today still struggle to optimise the allocation of orders, as many companies stick to simple processes to fulfil orders. However, today’s businesses operate in a global and dynamic environment. By using a more advanced system, you can bring many possibilities of improvement to this process.

You can use the concept of Available to Promise (ATP) to help with efficiently making the right order fulfilment decisions. ATP is a business function that provides a response to customer order enquiries based on availability of products. It generates information on available quantities of the requested product and delivery due dates. It represents the uncommitted portion of stock by looking at what is in stock and in transit, considering what is already promised to customers and calculating what is free to promise to your customers and when.

Example of a possible ATP set-up

Looking at an ATP set-up, the first “soft” confirmation can happen when the order comes in. It provides a first estimate of the ability to fulfil the customer demand. Orders can be re-confirmed according to allocation rules such as customer priorities and customer preferences (delivery in full or separate deliveries allowed etc.) at a later stage to provide a more “firm” confirmation. Using this type of set-up, you can increase the profitability of your order fulfilment decisions by taking into account customer profitability and priorities in allocating orders instead of just basing order allocation on sequence of incoming orders.

However, the cost of doing so is a delayed ”firm date” and quantity confirmation to your customers. In this case, the client cannot rely on the initial estimate but will have to wait until the batch allocation has run before they receive the reliable estimate. This is a trade-off that needs to be made considering the needs of the company and the industry or context that the company finds itself in.

For a company that operates in a context with a horizon of multiple weeks or months between ordering date and delivery date, using nightly batch allocations and thus a maximum of 24 hours during which the confirmation is not yet fixed, the trade-off for using batch allocation is much smaller than for companies operating in an environment of almost immediate delivery after order creation due to short lead times.

Example

Let’s take a simple example: On Monday morning, three orders are received – all with a requested delivery for delivery date 1.

  • At 10:00, an order for 5 pieces is received with a priority of 3 assigned.
  • At 11:00, an order for 5 pieces is received with a priority of 3 assigned.
  • At 13:00, an order for 3 pieces is received with a priority of 1 assigned.
Table 1. A simple example: Three orders come in on Monday for delivery date 1

While the unallocated portion of the stock looks like this:

  • For the first delivery date, only 10 items are unallocated in stock
  • For the second delivery date (due to a replenishment), 2 pieces are now unallocated
  • For the third delivery date, 5 pieces are unallocated in stock
Figure 3: Unallocated stock outlook

Based on this stock outlook and sequence of ordering, the confirmation will look like this:

Table 2. A simple example: At the end of Monday, the delivery confirmation looks like this
Monday 10:00

Order (#1) gets a soft confirmation for the requested delivery date (delivery date 1). Out of 10 unallocated pieces for delivery date 1, 5 pieces in stock are now allocated to order #1, and the stock outlook is updated to only 5 items being unallocated for delivery date 1.

Monday 11:00

Order (#2) gets a soft confirmation for the requested delivery date (delivery date 1). Out of 5 unallocated pieces for delivery date 1, 5 pieces in stock are now allocated to order #2, and the stock outlook is updated to no more items being available for confirmation for delivery date 1.

Monday 13:00

Order (#3) gets a soft confirmation for delivery date 2 and 3 since no more items are available for delivery date 1. Out of 2 unallocated pieces for delivery date 2, 2 pieces in stock are now allocated to order #3, and the stock outlook is updated to no more items being available for confirmation for delivery date 2. Out of 5 unallocated pieces for delivery date 3, 1 piece in stock is now allocated to order #3, and the stock outlook is updated to only 4 items being available for confirmation for delivery date 3.

Figure 4. Whenever an order comes in, a soft confirmation happens based on unallocated stock

During the day, the ATP has confirmed the orders sequentially rather than according to priority rules, resulting in the “soft” confirmation seen above in table 2.

Monday evening/night

During the night, the ATP is re-run, and the confirmation is updated in a batch job, meaning that the orders that were confirmed during the day are un-confirmed, and the unallocated stock situation will be the same as Monday morning. During the nightly batch job, the ATP will take the priorities into account and give a firmer confirmation. The new situation will look like this:

Table 3. A simple example: During the night, the delivery confirmation looks like this.
Priority 1

Order (#3) is prioritised and put first in line for the confirmation due to the assigned “priority 1”. The order gets a firm confirmation for the requested delivery date (delivery date 1). Out of 10 unallocated pieces for delivery date 1, 3 pieces in stock are now allocated to order #1, and the stock outlook is updated to only 7 items being unallocated for delivery date 1.

Priority 3

Both order #1 and order #2 have the same assigned priority, which is priority 3, so they are confirmed in sequence.

Order #1 gets a firm confirmation for the requested delivery date (delivery date 1). Out of 7 unallocated pieces for delivery date 1, 5 pieces in stock are now allocated to order #1, and the stock outlook is updated to only 2 more items being unallocated for delivery date 1.

Order #2 gets a firm confirmation partly for the requested delivery date (delivery date 1) as well as for delivery date 2 and 3. Out of 2 unallocated pieces for delivery date 1, 2 pieces in stock are now allocated to order #2, and the stock outlook is updated to no more items being available for confirmation for delivery date 1. Out of 2 unallocated pieces for delivery date 2, 2 pieces in stock are now allocated to order #3, and the stock outlook is updated to no more items being available for confirmation for delivery date 2. Out of 5 unallocated pieces for delivery date 3, 1 piece in stock is now allocated to order #3, and the stock outlook is updated to only 4 items being available for confirmation for delivery date 3.

Figure 5. During the night, the orders are re-confirmed based on priority

The ATP represents the “uncommitted portion of stock” as it considers what is in stock, in transit and what has already been promised. It then calculates what is free to promise to your customers and when. This uncommitted portion of stock can be represented in different ways, based on the needs of the company. This could be by only considering stock on hand or also considering “virtual stock” by taking into account (firm) planned orders and/or purchase orders.

Regarding the allocation of orders, you can also set up different priorities in the system based on which allocation can happen, such as customer priority, customer profitability, product classification, etc. Some customers might prefer only delivery in full, whereas others might accept partial deliveries spread out over time. The ATP can take such preferences into account.

Characteristics of an ATP

The ATP functionality has the authority to say yes to customer orders either on requested due date or with postponed delivery, but it cannot deny an order. If the ATP finds availability when it searches for uncommitted stock (or future supply), it accepts the order and reduces the uncommitted stock accordingly. If it can only find enough to fulfil the order partially, it reacts based on rules set up beforehand and either signal a problem or accept and confirm across different delivery dates. If no availability can be found, the system signals a problem, and the order is either manually rejected or manually confirmed.

The ATP system serves as an automatic decision-making tool in case no issues are present and allows manual exception-based handling as it will prompt for manual action in case a demand cannot be covered. So, the system is responsible for providing the right information in order to make informed decisions after careful evaluation and allows you to make time for the things that matter most.

In case the ATP signals that the order cannot be delivered according to the requested quantity, price or due date, it is up to the company to decide if there are options that make it possible to still deliver to the customer. You need to ask the following questions: Can substitute products be delivered on the requested due date? Can we deliver part of the requested amount on the initial requested delivery date and the rest later? Can we deliver in full later than the requested delivery date? Can we deliver from other sites or warehouses?

Once a customer order has been promised, the next step is to ensure that it will be ready at the warehouse and loading dock for shipment.

 

Where to start with your implementation?

If you have not started the journey to ATP yet, the first step is to move away from a First-In-First-Out approach and towards sorting orders by required shipping date and treating production and handling processes according to this sorting. To follow up on this way of working, we advise that you start with simple reports and KPI tracking to ensure correct handling of orders.

Once you have the right processes and KPIs in place, you could start by implementing a tool for a site or factory-based ATP that takes each site or factory into account as its own entity. Many planning systems now have the possibility of implementing an Available-to-Promise check, but before you can implement this option, it is important that you consider the prioritisation or allocation rules that are important to your business as a whole and to your customers. Identifying these and setting them up in the system is a key prerequisite to reap the benefits of the ATP.

If you are working in a global environment with locations across the world, the next step could be to move to a global solution. To support this, you could use a global ATP solution, which takes more than just the customer-facing location into account. In case you have multiple warehouses, sites or factories, their inventory or production capacity can be taken into account when calculating the ATP and can even provide the customer with a choice on what to do. The tool will search throughout the entire network and see if there is unallocated stock, and if or when it could be delivered to the right customer.

You are always welcome to reach out to discuss the application of these concepts in your organisation.

References

Wallace, T. F., & Stahl, R. A. (2003). Master scheduling in the 21st century: For simplicity, speed, and success ; up and down the supply chain. Cincinnati, OH:Wallace.

Peerless Research Group. (2016). From Customer Orders Through Fulfillment: Challenges and Opportunities in Manufacturing, High-Tech and Retail (Publication). Retrieved from https://www.supplychain247.com/paper/from_customer_orders_through_fulfillment