Monday, September 3, 2018

VDML for Business Architects: Part 10 of 11, Agile Business Architecture

Please see the post for VDML for Business Architects: Part 1 of 11, for the introduction to this series of posts.  This part describes a recommended business architecture supported by VDML.

1.    Service oriented architecture

VDML is designed to support a service-oriented architecture with services defined by collaborations, particularly capability methods.  However, VDML can also model a silo-oriented architecture and facilitate transformation to a service-oriented architecture by carving out and consolidating capability methods.
2.    Business partner ecosystem

Business network models support analysis of relationships with business partners such as exchange of value propositions and deliverables.  In successful relationships, each participant must realize adequate benefit to sustain the relationships.
3.    Business function outsource modeling

Outsourcing moves business capabilities to services provided by an external business partner.  The VDML business model provides the basis for identification of operations to be transitioned and the points where interactions must occur to engage and oversee the outsourced capabilities.
4.    Meta-methods operate on mainstream operations
 
Certain capabilities operate on the elements of the mainstream business.  For example, maintenance and repair services operate on machines.  Personnel services manage personnel records and benefits.  Financial services manage billing, payments and inventory records and transactions.  Strategic planning activities analyze current circumstances and evaluate potential strategies, transformations or improvements.  Such operations can be modeled in VDML with elements of the mainstream operations represented as business items.

VDML for Business Architects: Part 11 of 11, Business Transformation Extension


Please see the post for VDML for Business Architects: Part 1 of 11, for the introduction to this series of posts.  This part describes the extension of VDML to support business transformation planning and management.
Failure to achieve and maintain a shared understanding of a business transformation is a major cause of failure to achieve expectations.  VDML supports development of a consistent business design that ensures a clear understanding of intent and requirements across affected organizational units.
2.    Incremental transformation phases (proposed VDML extension)
Incremental transformation phases are represented as VDML model scenarios with the expectation that each scenario represents a viable operating model as an interim step in the transformation.  Transformation work is identified based on the changes from one phase scenario to the next.
3.    Transformation details (proposed VDML extension)
Transformation requirements must be evaluated based on the input/output requirements of the changed capabilities. Changed capabilities must be analyzed to determine the requirements for other source or recipient capability methods.
Oversight of transformation progress is supported by key objectives. Of primary interest are objectives to achieve the next phase of the transformation plan.  Objectives identify the responsible manager and persons with an interest in progress on the objective.
Changes required to capability methods, resources and organizations will have impact on related capability methods and organizations linked through deliverable flows and delegations.  These consequential changes must also be identified in the transformation plan for each phase.
The transformation plan identifies changes by phase, and the associated scenarios determine dependencies and responsibilities affected by those changes.  Consequently, objectives for each phase can be generated for the required changes along with dependencies between objectives based on the requirements for completion of each phase of the transformation plan and dependencies based on deliverable flows within the phase.

 

Monday, August 27, 2018

Multiple Vocabulary Facility Progress Report

Since my post in April, 2016, there have been submissions to the OMG (Object Management Group) RFP (the RFP is available at https://www.omg.org/cgi-bin/doc?ad/16-03-04 ).  The goal is to define a standard facility that can be used to implement multiple vocabulary capabilities in implementations of OMG modeling languages.  The submitters have joined together to develop a single specification and submitted a joint draft on February 19, 2018.  There is agreement on a core structure that will enable a modeling language user to select a preferred vocabulary for terms that appear in that user's interactive interface.

However, there are some unresolved issues regarding the details of the specification.  For example, in a single vocabulary, the same term may be used to represent different elements in the same model.  In a business process model, an activity name may be used for different activities in different processes.  The meaning of the activity name depends on the process context.  Thus, a term may represent different concepts in different contexts.  This is fairly common in programming languages and models.  This is a characteristic of a modeling language that affects the representation of elements in the models it creates. Thus restricts the user's selection of terms new elements are added to a model, and it restricts another user who builds another vocabulary to represent the concepts of the model for a different community of users.

There are also issues regarding the integration of an MVF facility with a modeling language implementation.

A new draft specification is expected for the December, 2018, OMG meeting.  OMG makes draft specifications available to OMG members, but does not make specifications publicly available until a final submission is adopted.

Thursday, April 21, 2016

Support for Modeling in Multiple, Alternative Natural Languages


For a number of years, I have looked for a way to enable implementations of OMG modeling languages to support multiple, alternative natural languages.  It should be possible to develop a model with a modeling tool that expresses modeling language elements in your native language, allows you to identify elements in your model in terms from your native language, and allows you to express definitions for your terms in your native language.  Then, it should be possible to give your model to a person with a different native language and have the modeling language and your model expressed in terms with definitions from that person’s native language. Of course, somebody must do the translation, but over time, there should be shared libraries of terms and definitions for the same concepts in different languages at least for certain industries or business disciplines.
I have made several attempts to initiate a request for proposals (RFP) in the Object Management Group (OMG) to solicit solutions to this capability.  It potentially expands the market for OMG modeling languages, and it would be particularly beneficial to organizations with operations in multiple countries.  Furthermore, it would shift the focus of creation of modeling elements from debates over a name to development of clear definitions of concepts.  The alternative languages might not only be languages of different countries, but languages (vernacular) of different disciplines.
At the March meeting of the OMG, we finally issued a Request for Proposals for a Multiple Vocabulary Facility.  I developed a draft several meetings ago, but Evan Wallace of NIST (US National Institute of Standards and Technology) brought insights from his work on semantics and ontologies to clarify the specific need and distinguish it from other specifications that have some forms of multiple language facilities.
The intent is that a modeling language implementation can integrate an MVF product and access libraries of terms and definitions to be integrated into the user inputs and displays.  This allows the user models to retain the capability if moved to a different modeling language tool as long as it supports MVF.
Submitters to this RFP must be OMG members at an appropriate membership level (Contributing or Platform), but submissions are often the product of coalitions with at least one qualified submitter.  Letters of intent are due to OMG by August 29, 2016.  Initial submissions are due February 20, 2016.  The process allows for submitters to reconcile differences after initial submissions for the possibility that they can converge to one, broadly accepted final submission.

Thursday, January 22, 2015

Exploring VDML with Spreadsheets

I developed a limited implementation of VDML (Value Delivery Modeling Language) using spreadsheets to illustrate the concepts and develop a better understanding of the value computations.  The document describing this implementation and a companion Excel workbook are available at the VDMBee web site: http://vdmbee.com/2015/01/exploring-vdml-spreadsheets/ 

I would be interested in feedback, particularly on VDML since the spreadsheet implementation is essentially a prototype, but the technique can be used for a limited application of VDML.

For more on VDML, see my overview at http://fredacummins.blogspot.com/2013/02/value-delivery-modeling-language-vdml_4878.html

Saturday, September 20, 2014

Does a company have a capability if the capability is outsourced?


In a LinkedIn discussion at https://www.linkedin.com/groupItem?view=&item=5912670423937478660&type=member&gid=84758&trk=eml-b2_anet_digest-hero-4-hero-disc-disc-0&midToken=AQFNSRpfQ-416Q&fromEmail=fromEmail&ut=1PAPLXT2WvyCo1 the question was raised, if a capability is outsourced, is it still a capability of the outsourcing company?  Capability analysis has gained attention from capability maps.   Capability also is a key concept of the Value Delivery Modeling Language (VDML), see http://fredacummins.blogspot.com/2013/02/value-delivery-modeling-language-vdml_4878.html .  Capabilities are fundamental to a high-level understanding what a company has or requires to deliver its product or service.  I think the LinkedIn discussion suggests that we do not have a shared definition of outsourcing.
To me, outsourcing is engaging an independent business entity to provide a commodity product or service. The outsourced operations represent a required capability, but the implementation is a black box to the purchaser.  The outsource provider is free to perform/implement the capability as it determines best to meet its obligations on delivery of the product or service and be competitive.  

Consider that the outsource provider may, in turn, outsource some of its needed capabilities.  Furthermore, the outsource provider should also be free to provide the same capability to other purchasers/consumers, including competitors of the purchaser—that is one way it can achieve economies of scale not available to its consumers.  The provider also will try to develop its capability in a way that gives it a competitive advantage over its competitors.
The product or service must be a commodity because the purchaser cannot put its key business capabilities in the hands of an external source to be shared with competitors.  That provider might not have the capacity to meet demand of increased business of the purchaser.  Furthermore, the provider may decide it is advantageous to devote increased capacity to the purchaser’s competitor.  If the provider fails to perform, then what? 

The purchaser must be able to find an alternative source.  Of course the provider may fail as a business, and the purchaser must be able to obtain the product or service elsewhere.  The purchaser should obtain the outsourced services through competitive bidding which ensures appropriate pricing as well as the availability of alternative providers.  The purchaser might be well advised to obtain the product or service from multiple providers. 
If a company needs to engage in defining or managing the details of an outsourced operation, then the capability should not be outsourced.  However, the company does need to invest in managing the relationship with the provider to ensure that the provider meets expectations such as product specifications, responsiveness, product quality, risks, and security that might adversely affect the company’s business.

If we are not dealing with a commodity product or service, I would characterize the arrangement differently.  For example, a franchise involves engaging outside business entities to provide necessary capabilities, but the franchiser will generally require an exclusive franchisee, and control over much of the operation so that there is consistency in the end product or service, and multiple franchise operators (as well as the franchisee) can benefit from economies of scale achieved by the franchiser.
If the purchaser needs critical capabilities for a new business opportunity, it may not be able to develop the capability, or it may not have the necessary intellectual property to be competitive.  Then the needed product or service is no longer commodity.  The purchaser cannot simply engage a black box, and it is desirable to have an exclusive arrangement to exclude competitors.  A joint venture is a possible arrangement.  The participants recognize their interdependence, and they make commitments to working together.  Typically there will be shared benefit from the success of the joint enterprise.

Engaging contract personnel might be considered another form of outsourcing.  Contract personnel may be engaged to increase capacity but with their activities restricted to protect key activities.  However, the company may obtain important special skills from contract personnel, and may more directly control the work products and intellectual property used or created by the contract personnel to preserve competitive advantage.  However, this is a weaker form of possessing the key aspects of an essential capability.
This might be a good time to consider the difference between a capability and a competency.  When a business has or develops a competency, that means it controls the key capabilities to be competitive, or realize competitive advantage.  This includes possession of the key resources, intellectual property, relationships, etc.  An enterprise may obtain a capability through outsourcing, but it cannot realize a competency through outsourcing.

So a capability may be outsourced if it is a commodity and outsourcing achieves economies of scale and/or relieves the outsourcing company of the burden and risks of managing an internal capability.  It is still a capability that is part of the company business.  A competency cannot be outsourced.  There are other alternatives to realizing capabilities from external sources that are essential to a competency, but these involve more collaborative and exclusive relationships that protect the competency and preserve competitive advantage.

Sunday, December 29, 2013

VDML Support for Balanced Scorecards and Strategy Maps


The Kaplan and Norton Balanced Scorecard(BSC) and Strategy Map (SM) are well-known, abstract models for structuring business transformation objectives.  The analysis of objectives helps refine a strategic plan, and the resulting objectives provide a basis for continuing assessment of progress.  This article describes how VDML can provide more robust analysis for setting BSC objectives and defining SM relationships.  VDML (Value Delivery Modeling Language) is a specification in the final stages of adoption by the OMG (Object Management Group).

BSC/SM objectives represent desired changes to the state of the business—key measurements of the operation of the business will drive improvements.  Thus the changes represented by BSC/SM objectives represent key differences between VDML As-Is and To-Be models. 

The diagram, below, characterizes the objectives of a BSC and SM.  For the BSC, Kaplan and Norton define four perspectives that classify objectives: (1) Learning and Growth, (2) Internal Processes, (3) Customer Value, and (4) Financial.  This classification drives a broader analysis starting with the development of capabilities (Learning and Growth), through development of internal methods and processes (Internal Processes), to delivery of customer value (Customer Value), and finally to enterprise success and sustainability (Financial).

 

The Strategy Map identifies causal relationships—some objectives depend on achievement of other objectives.  For example, an internal process will not be successful without the capabilities that support it (e.g., people, machines, knowhow).  The diagram illustrates the general flow of causal relationships—an actual strategy map would link specific objectives.

A VDML model can represent the current or future state of the business and associated operating measurements.  A model could involve multiple lines of business and multiple value streams.  VDML models can extend the BSC/SM analysis to support more robust transformation planning and assessment.  The diagram, below, depicts this.


So a VDML To-Be model can represent the improvements in performance and value creation if the strategy is successful.  As the basis for objectives, these To-Be measurements must be compared to corresponding measurements in an As-Is (current state) model.  The current As-Is model will be updated to reflect changes as the transformation progresses.  If the transformation goes as planned, the As-Is model will become the same as the To-Be model (transformation complete).  Development of these models is discussed in greater detail in Strategic Planning with VDML.

VDML can represent many of the measurements of interest for the BSC/SM objectives, so a BSC/SM objective may obtain its target measurement from the To-Be model and its current measurement from the As-Is model.  VDML does not directly identify objectives for Learning and Growth, but can represent the requirements for resources that must be developed to achieve capability requirements.

VDML can represent objectives of the BSC Financial Perspective as As-Is and To-Be value proposition measurements and market segment forecasts.  However, VDML does not provide support for the development of these market forecasts.  For example, profit is essentially price minus cost.  VDML supports cost detail, but price is a management decision based on market analysis.  Market share is a forecast based on price and other factors including future competition.  VDML can capture those measurements, but does not provide support for market analysis. 

Operational measurements support more extensive analysis and mapping.  For example, if a new capability is needed, the need is represented by the absence of the capability in the As-Is model.  The objective measurement is essentially binary.  If the capability requires skilled people, we might represent the need for growth with a VDML store of people required for the capability in the As-Is and To-Be models.  The To-Be model may also represent new supplier relationships.

From a VDML perspective, causal relationships (i.e., dependencies) are of six different types: (1) changes to enterprise capabilities (i.e., “capital” in Learning and Growth) that are necessary for implementation of Internal Perspective value stream changes, (2) changes to capabilities that increase customer value (Customer Perspective), (3) changes to capabilities that improve investor value (Financial Perspective), (4)  changes to a value stream that affect activity value contributions to customer value (Customer Perspective) that may involve complex changes reflecting multiple dependencies between activities, (5) changes to the value stream that improve investor value (Financial Perspective), and (6) changes in customer satisfaction levels that drive market changes reflected in investor value (Financial Perspective).  These can be seen in the causal relationships depicted in the first diagram, above.

While these are all supported by VDML, they require two VDML models—the As-Is and a To-Be models.  Type (1) can be observed by a change in capabilities (new or modified) between the As-Is and To-Be models.  Type (2) can be observed as value stream changes that improve desired customer value proposition measurements.  Type (3) requires human recognition of the value investors will place on improved enterprise capabilities.  Type (4) can be observed by tracing the sources of changes to the customer values in the To-Be model to the relevant customer value contributions that are improved from the As-Is model.  Type 5 can be observed from changes in value contributions (such as cost reductions) that have a direct impact on investor value.  Type 6 is based on changes in the customer value propositions but it requires market insight and feedback regarding trends and competition to determine the impact on investor value.  In most cases, these causal relationships do not correspond directly to VDML deliverable flows, but rather they represent the consequential effects of changes to As-Is measurements that will directly or indirectly result in changes to To-Be measurements. 

In a substantial transformation, there must be phases of implementation.  Without objectives (or intermediate target objectives) for phases, management would have difficulty assessing progress on a long-duration project with multiple components.  Each phase should be represented by a VDML model for the expected state of the business at the end of the phase.  The development of a transformation plan and associated VDML models for transformation phases is discussed in greater detail in Strategic Planning with VDML.  As the transformation progresses, a series of As-Is models would be created to represent the incremental, current state of the business. The As-Is model evolves toward the To-Be model so that measurements in the current state depict progress toward To-Be targets. 

The BSC/SM model then becomes a series of phase models with As-Is and To-Be supporting models to define intermediate targets—see the diagram, below.  The To-Be VDML model of a phase is the expected As-Is model of the next phase for planning purposes, but the actual As-Is model for a “next” phase will likely vary some from the planned To-Be model.  A BSC/SM short-term, current-phase model represents the objectives of the current phase.  The full BSC/SM model can be derived from the initial As-Is model and the final To-Be model. 


While the full BSC/SM model can be stable (unless the end state of the business evolves), the current-phase BSC/SM model will be incrementally based on new versions of the As-Is model as the transformation progresses, and it will be based on new To-Be models as new phases are started.

In summary, VDML can provide significant value in validating the strategy, developing the transformation plan, and setting and evaluating objectives and causal relationships appropriate to each phase of transformation.  However, the relationships between the BSC/SM objectives and the VDML models require human input to select appropriate key objectives and identify the causal relationships.  Modeling mechanisms for a BSC/SM extension are not defined in the current VDML specification, but are left for VDML implementers to develop if they determine that there is sufficient market demand.