Why does prototyping make sense?
“Measure twice; cut once.”
This old adage is relevant to more than just Carpentry. Automotive companies wouldn’t build a car without first prototyping. Prototyping allows for a ‘hands-on’ feedback opportunity in order to correct errors before going to product manufacture. Almost all industries that produce products routinely use prototyping. One of the exceptions is the construction industry, which is used to having to create ‘one off’ designs in the absence of a prototype model. In fact one of the well known issues facing cost control in the construction industry is:
“A lack of prototype development resulting in untested and ill specified components and technologies”
(From – Willis’s Practice and Procedure for the Quantity Surveyor – Allan Ashworth, 12th Edn)
3D modelling is going some way towards addressing this in the construction industry. Quantity surveyors are concerned with reducing project cost risk in a labour intensive industry full of ‘one off’ projects. There are parallels with industrial automation software design, except there has not been a 3D modelling equivalent, or even a data modelling equivalent, until now.
A software ‘mock up’ or architectural ‘spike’ is essentially a software prototype which allows more user collaboration and greater crystallisation of requirements at an earlier stage in an automation software design project.
The chances of expensive software change control costs, further down the project execution timeline, rise rapidly if user requirements are not captured prior to commencing detailed software design. Every process manufacturing vendor that has embarked on a greenfield project expansion is sensitive to this concern.
The project costs with least certainty tend to be the manufacturing control system development costs. The traditional approach is to get three prospective vendors to bid, based on a general user requirement specification, and then ‘bid level’ and eventually chose a winner. Differences of 50% are not unheard of, between the bids (in the authors’ experience). This traditional approach does not sufficiently reduce the risk of scope creep or change control throughout project design. (The client is basically taking the prospective vendors solution without sufficiently challenging it. A process akin to writing the vendor a blank cheque).
The only winner here is the automation system integrator. In order to win the tender in the first instance, the integrator will have ‘low-balled’ their quote. As a result, they look to improve the bottom line at every juncture through change control costs throughout the project life-cycle. This makes for a prickly client-software supplier relationship throughout the project. No automation manager wants to go to his boss looking for additional funding as a result of miscalculating software burn costs. That’s a career limiting conversation.
The only way to reduce automation software project cost/scope/schedule risk is with better FEED (Front End Engineering Design) and this is where the prototyping of the entire factory ISA-S88/S95 recipes, procedures and data model fits in.
ISA-S88 prototyping
In pharmaceutical batch manufacture environments, R&D departments would typically develop ISA-S88 general recipes and provide this information to production and engineering. This information is normally translated into equipment specific S88 recipes. Spike doesn’t do this for you, but rather provides a framework to quickly map out the requirements, in a virtual control system environment.
The ISA-S88-compliant physical and procedural model simplifies and accelerates process modelling, design, training and testing. In Spike, the batches can be run virtually too. Users can run batches from their desk to see how the equipment operation changes over time, see what the batch report is going to look like, and get a better feel for system operation.
ISA-S95 & MES eMBR design prototyping
ISA-S95 is in general, focused on exchange of information between layer 3 (MES) and layer 4 (ERP) systems, whereas ISA-S88 focuses on actual batch control.
Spike provides an SFC workflow and programming environment for eMBR (electronic master batch record) design. It enables users to design electronic Batch Records (Recipes), including user interactions with prompts, messages and forms. Resource classes (material, equipment ,personnel) can be documented, and Bill Of Material recipes can be created. Material Flow Control is catered for by allowing user define input, output and transfer material parameters per SFC step. Because Spike has its own Batch workflow engine, eMBRs can be simulated before final design specification export.
ISA-S88 / ISA-S95 Interaction
Critically, Spike allows the user emulate the interaction between batch control systems and MES systems by means of ISA-S95 style transaction modules. These modules can then be exported and used as the basis of an integration specification. The interactions can be demonstrated live to prospective users and internal clients alike, to ensure everyone is on the same page. The integration of control systems with MES systems is a key headache flashpoint for a number of reasons:
- Different software vendors for batch system, and MES system.
- Different software architecture nomenclature. ISA-S88 & ISA-S95 terminology and approach simply do not mesh well together. Their database schemas overlap awkwardly.
- Different project execution priorities (Get the batch system commissioned then bolt on the MES system afterwards).
- Different internal project execution teams often with misaligned goals.
Prototyping the entire S88/S95 interaction upfront can alleviate a lot of this pain. Spike enables this, and also simplifies the ISA-S95 terminology to make training and understanding easier.
Collaboration is easier as Spike is web server based (unlimited clients). Finally, once the team is satisfied with the eMBR behaviour, the entire configuration can be exported as a design specification. Or even PLCopen format for direct import to an open standards compliant target control system. (But that is version 2.0…)
Accelerate your design project execution life-cycle through collaborative interactive electronic batch design reviews.
Accelerate your Automation project design with Spike
Comments are closed.