You’re looking to purchase a configurator. You do some research and find that experts say that you have two choices for your solution: rules-based configuration or constraint-based configuration. We’ll start this blog by saying that we might be experts about configuration, but when it comes to explaining these two types of checks and balances, we need to take a deep breath. We’d like to describe rules and constraints, and explain how they help product modelers and salespeople create feasible, accurate offerings. Our goal is to help you understand which overall approach is best for your business. We’d like also to show you how flexibility is more effective than rigid rules and constraints–combining the two approaches in one solution might be the best solution
Constraints represent overarching restrictions for potential product combinations or choices. They indicate “what you want to do” rather than “how to do it.” Each constraint is referred to as a single “statement.” Constraints establish the contours of rules patterns and guide primary configuration activities, making it easier to develop overarching elements of configuration models over time.
How Constraints work: configuring a door
For this example , we are working with 3 statements, or 3 constraints.:
- Door (Height,50) when Height < 2200.
- Door (Height,Thickness) when Height == 2200, Thickness >= 60.
- OuterDoor (Height,Thickness) when Thickness > 80, Door(Height,Thickness).
The constraints establish a pattern of rules for how the door can be configured. If a different height is chosen, that pattern will remain constant, even if the variables change.
Here’s an example of what goes on behind the scenes if a customer wants a door that is a different thickness—say 90 mm. These are the constraints that will be guiding the configuration:
Customer wants Outer Door (Height,90)
Example of constraint satisfaction:
Outer Door (Height,90) is only satisfied when 90 > 80, Door (Height,90).
That results in goal Door (Height,90).
Door(Height,90) is only satisfied when Height == 2200 and 90 >= 60.
That results in Door(2200, 90)
Rules-based methods for configurator software describe logical progressions for baseline configurations. This approach anticipates all possible ways to create finished products, services, and deals. For sales people, rules-based configuration sets parameters for all choices, using restrictive statements that specify precisely what to do in every scenario. The rules take salespeople through sequences of prescribed selling for configuring a product—including all options, conditions, and actions. If a sales person enters an option that does not fit into a sequence, additional rules prompt them to enter a viable option.
Comparing with the example descripted above, here the same scenario for Rules-Based Configuration:
- If height < 2200 then Thickness = 50 end if
- If Thickness > 60 then Height = 2200 end if
- If Thickness > 80 then DoorType = Outerdoor end if
Comparing Constraint-Based and Rules-Based Configuration
Rules can be complex, since they require scripting language, but a rules-based approach offers flexibility for tailoring visibility, creating defaults, adding text, and other options. Constraint-based configuration offers simplicity but limits ability to work with details such as visibility, defaults, text—the options that a rules-based approach offers. e-Con is all about flexibility, but we’re also committed to making it easier to sell and produce unique offerings. So we’re moving beyond comparing to combining constraints and rules.
Configure more options, better, faster by combining constraints and rules
Companies can realize great benefits from both approaches, so the e-Con Sales and Product Configurator is built on a rules-based engine that also offers decision trees for constraint-based configuration.
Though we refer to e-Con as a rules-based configurator, we’ve enhanced that approach with 4 capabilities to give companies a tool with more value than one that is strictly rules-based or contraints-based:
- We’ve made it possible to group rules in e-Con. Rule grouping makes it easier to maintain complex configuration models. You can group all rules for a certain task together in a model to make it easy to find where some functionality of the model has been implemented.
- Users can create sub-models, making it possible to package complex data in logical groupings—for example, by using encapsulation, information hiding, or reuse of sub-models.
- We’ve also developed a DMI tool for maintaining e-Con installations and environments. In addition, users can work within the e-Con Studio to be able to log and debug rules, sharply reducing the need for deep IT knowledge and requests for third-party assistance. These tools ensure:
- Fewer failures and requests for support during installation of the configurator
- Reduced error costs for customers, partners, and ISVs
- Increased technical support that doesn’t add overhead
- Less risk and less time needed to set up different environments—with DMI, error-free environments can be set up in minutes
- We’ve also made use of a constraint-based approach by adding wizards for rules, making it easier to set up complex configuration models and giving modelers optimal flexibility. In short, with a wizard you’re able to very easily and visually set up of decision matrices for constraints with two variables. With a decision tree you can set up constraints with more than two variables, both are easy to set up and maintain. See example below:
With our example of e-Con, we show you how adding critical, holistic constraints can optimize rules-based processes. Our suggestion? Of course we want you to explore the enhancements that create a bullet-proof configurator tool. You’ll find great hand’s on models to work with at our demo-lab. But we also encourage you to ask your own questions about ‘what’ and ‘how’ right here.
Rules and constraints are packed with topics for discussion, so share you voice below!Product groups, production teams, salespeople, ordering and procurement—virtually all teams working with your internal product cycles will tell you that they are confident about the nuances of ‘what’ has been configured as well as ‘how’ to configure it.’