The whole of enterprises’ endeavors and behaviors cannot be coerced into models lest they inhibit their ability to navigate ill defined and shifting business environments. Enterprises immersion in digital environments is making limits all the more explicit:
On the environment side, facts, once like manna from heaven ready to be picked and interpreted, have turned into data floods swamping all recognizable models imprints
On the symbolic side, concepts, once steadily supported by explicit models and logic, are now emerging like new species from the Big Data primordial soup.
Typically, business analysts are taking the lead on both fronts toting learning machines and waving knowledge graphs. In between system architects have to deal with a two-pronged encroachment on information models.
On the one hand they have to build a Chinese wall between private data and managed information to comply with regulations
On the other hand they have to feed decision-making processes with accurate and up-to-date observations, and adjust information systems with relevant and actionable concepts.
That brings a new light on the so-called conceptual, logical, and physical “data” models as key components of enterprise architecture:
Physical data models are meant to be directly lined up with operations and digital environments
Logical models represent the categories managed by information systems and must be up to par with systems functional architecture
Conceptual models are meant to represent enterprise knowledge of business domains and objectives, as well as its embodiment in organisation and people.
Logical models (information) appear therefore as an architecture hub linking business facts (data) and concepts (knowledge), ensuring exchanges between environments and representations e.g.:
Knowledge analysing data looking for new business categories (a)
Knowledge using business categories to build new data sets (b)
Data being crossed with business categories to improve knowledge (c)
Taking advantage of the digital transformation, such exchanges can be turned into osmosis between systems and information architectures.
The digital transformation induces fundamental changes for the exchanges between enterprises and their environment.
To begin with, their immersion into digital environments means that the traditional fences surrounding their IT systems are losing their relevance, being bypassed by massive data flows to be processed without delay.
Then, the induced osmosis upturns the competition playground and compels drastic changes in governance: less they fall behind, enterprises have to redefine their organization, systems, and processes.
Strategic thinking is first and foremost making differences with regard to markets, resources and assets, and time-frames. But what makes the digital revolution so disruptive is that it resets the ways differences are made:
Markets: the traditional distinctions between products and services are all but forgotten.
Resources and assets: with software, smart or otherwise, now tightly mixed in products fabric, and business processes now driven by knowledge, intangible assets are taking the lead on conventional ones.
Time-frames: strategies have for long been defined as a combination of anticipations, objectives and policies whose scope extends beyond managed horizons. But digital osmosis and the ironing out of markets and assets traditional boundaries are dissolving the milestones used to draw horizons perspectives.
To overcome these challenges enterprises strategies should focus on four pillars:
Data and Information: massive and continuous inflows of data calls for a seamless integration of data analytics (perception), information models (reasoning), and knowledge (decision-making).
Security & Confidentiality: new regulatory environments and costs of privacy breaches call for a clear distinction between data tied to identified individuals and information associated to designed categories.
Innovation: digital environments induce a new order of magnitude for the pace of technological change. Making opportunities from changes can only be achieved through collaboration mechanisms harnessing enterprise knowledge management to environments intakes.
The seamless integration of enterprise systems into digital business environments calls for a resetting of value chains with regard to enterprise architectures, and more specifically supporting assets.
Concerning value chains, the traditional distinction between primary and supporting activities is undermined by the generalization of digital flows, rapid changes in business environments, and the ubiquity of software agents. As for assets, the distinction could even disappear due to the intertwining of tangibles resources with organization, information, and knowledge .
These difficulties could be overcome by bypassing activities and drawing value chains directly between business processes and systems capabilities.
From Activities to Processes
In theory value chains are meant to track down the path of added value across enterprise architectures; in practice their relevancy is contingent on specificity: fine when set along silos, less so if set across business functions. Moreover, value chains tied to static mappings of primary and support activities risk losing their grip when maps are redrawn, which is bound to happen more frequently with digitized business environments.
These shortcomings can be fixed by replacing primary activities by processes and support ones by system capabilities, and redefining value chains accordingly.
From Processes to Functions & Capabilities
Replacing primary and support activities with processes and functions doesn’t remove value chains primary issue, namely their path along orthogonal dimensions.
That’s not to say that business processes cannot be aligned with self-contained value chains, but insofar as large and complex enterprises are concerned, value chains are to be set across business functions. Thus the benefit of resetting the issue at enterprise architecture level.
Borrowing EA description from the Zachman framework, the mapping of processes to capabilities is meant to be carried out through functions, with business processes on one hand, architectures capabilities on the other hand.
If nothing can be assumed about the number of functions or the number of crossed processes, EA primary capabilities can be clearly identified, and functions classified accordingly, e.g: boundaries, control, entities, computation. That classification (non exclusive, as symbolized by the crossed pentagons) coincides with that nature of adjustments induced by changes in business environments:
Diversity and flexibility are to be expected for interfaces to systems’ clients (users, devices, or other systems) and triggering events, as to tally with channels and changes in business and technology environments.
Continuity is critical for the identification and semantics of business objects whose consistency and integrity have to be maintained along time independently of users and processes.
In between, changes in processes control and business logic should be governed by business opportunities independently of channels or platforms.
Processes, primary or otherwise, would be sliced according to the nature of supporting capabilities e.g: standalone (a), real-time (b), client-server (c), orchestration service (d), business rules (e), DB access (f).
Value chains could then be attached to business processes along these functional guidelines.
Tying Value Chains to Processes
Bypassing activities is not without consequences for the meaning of value chains as the original static understanding is replaced by a dynamic one: since value chains are now associated to specific operations, they are better understood as changes than absolute level. That semantic shift reflects the new business environment, with manufacturing and physical flows having been replaced by mixed (SW and HW) engineering and digital flows.
Set in a broader economic perspective, the new value chains could be likened to a marginal version of returns on capital (ROC), i.e the delta of some ratio between value and contributing assets.
Digital business environments may also made value chains easier to assess as changes can be directly traced to requirements at enterprise level, and more accurately marked across systems functionalities:
Logical interfaces (users or systems): business value tied to interactions with people or other systems.
Physical interfaces (devices): business value tied to real-time interactions.
Business logic: business value tied to rules and computations.
Information architecture: business value tied to systems information contents.
Processes architecture: business value tied to processes integration.
Platform configurations: business value tied to resources deployed.
The next step is to frame value chains across enterprise architectures in order to map values to contributing assets.
Assets & Organization
Value chains are arguably of limited use without weighting assets contribution. On that account, a major (if underrated) consequence of digital environments is the increasing weight of intangible assets brought about by the merge of actual and information flows and the rising importance of economic intelligence.
For value chains, that shift presents a double challenge: first because of the intrinsic difficulty of measuring intangibles, then because even formerly tangible assets are losing their homogeneity.
Redefining value chains at enterprise architecture level may help with the assessment of intangibles by bringing all assets, tangible or otherwise, into a common frame, reinstating organization as its nexus:
From the business perspective, that framing restates the primacy of organization for the harnessing of IT benefits.
From the architecture perspective, the centrality of organization appears when assets are ranked according to modality: symbolic (e.g culture), physical (e.g platforms), or a combination of both.
On that basis enterprise organization can be characterized by what it supports (above) and how it is supported (below). Given the generalization of digital environments and business flows, one could then take organization and information systems as proxies for the whole of enterprise architecture and draw value chains accordingly.
Value Chains & Assets
Trendy monikers may differ but information architectures have become a key success factor for enterprises competing in digital environments. Their importance comes from their ability to combine three basic functions:
Mining the continuous flows of relevant and up-to-date data.
Analyzing and transforming data, feeding the outcome to information systems
Putting that information to use in operational and strategic decision-making processes.
A twofold momentum is behind that integration: with regard to feasibility, it can be seen as a collateral benefit of the integration of actual and digital flows; with regard to opportunity, it can give a decisive competitive edge when fittingly carried through. That makes information architecture a reference of choice for intangible assets.
Insofar as enterprise architecture is concerned, value chains can then be threaded through three categories of assets:
Turning thoughts into figures faces the intrinsic constraint of dimension: two dimensional representations cannot cope with complexity.
So, lest they be limited to flat and shallow thinking, mind cartographers have to introduce the cognitive equivalent of geographical layers (nature, demography, communications, economy,…), and archetypes (mountains, rivers, cities, monuments, …)
Nodes: What’s The Map About
Nodes in maps (aka roots, handles, …) are meant to anchor thinking threads. Given that human thinking is based on the processing of symbolic representations, mind mapping is expected to progress wide and deep into the nature of nodes: concepts, topics, actual objects and phenomena, artifacts, partitions, or just terms.
It must be noted that these archetypes are introduced to characterize symbolic representations independently of domain semantics.
Connectors: Cognitive Primitives
Nodes in maps can then be connected as children or siblings, the implicit distinction being some kind of refinement for the former, some kind of equivalence for the latter. While such a semantic latitude is clearly a key factor of creativity, it is also behind the poor scaling of maps with complexity.
A way to frame complexity without thwarting creativity would be to define connectors with regard to cognitive primitives, independently of nodes’ semantics:
References connect nodes as terms.
Associations: connect nodes with regard to their structural, functional, or temporal proximity.
Analogies: connect nodes with regard to their structural or functional similarities.
At first, with shallow nodes defined as terms, connections can remain generic; then, with deeper semantic levels introduced, connectors could be refined accordingly for concepts, documentation, actual objects and phenomena, artifacts,…
Semantics: Extensional vs Intensional
Given mapping primitives defined independently of domains semantics, the next step is to take into account mapping purposes:
Extensional semantics deal with categories of actual instances of objects or phenomena.
Intensional semantics deal with specifications of objects or phenomena.
That distinction can be applied to basic semantic archetypes (people, roles, events, …) and used to distinguish actual contexts, symbolic representations, and specifications, e.g:
Car (object) refers to context, not to be confused with Car (surrogate) which specified the symbolic counterpart: the former is extensional (actual instances), the latter intensional (symbolic representations)
Maintenance Process is extensional (identified phenomena), Operation is intensional (specifications).
Reservation and Driver are symbolic representations (intensional), Person is extensional (identified instances).
It must be reminded that whereas the choice is discretionary and contingent on semantic contexts and modeling purposes (‘as-it-is’ vs ‘as-it-should-be’), consequences are not because the choice is to determine abstraction semantics.
For example, the records for cars, drivers, and reservations are deemed intensional because they are defined by business concerns. Alternatively, instances of persons and companies are defined by contexts and therefore dealt with as extensional descriptions.
Abstractions: Subsets & Sub-types
Thinking can be characterized as a balancing act between making distinctions and managing the ensuing complexity. To that end, human edge over other animal species is the use of symbolic representations for specialization and generalization.
That critical mechanism of human thinking is often overlooked by mind maps due to a confused understanding of inheritance semantics:
Strong inheritance deals with instances: specialization define subsets and generalization is defined by shared structures and identities.
Weak inheritance deals with specifications: specialization define sub-types and generalization is defined by shared features.
The combination of nodes (intension/extension) and inheritance (structures/features) semantics gives cartographers two hands: a free one for creative distinctions, and a safe one for the ensuing complexity. e.g:
Intension and weak inheritance: environments (extension) are partitioned according to regulatory constraints (intension); specialization deals with subtypes and generalization is defined by shared features.
Extension and strong inheritance: cars (extension) are grouped according to motorization; specialization deals with subsets and generalization is defined by shared structures and identities.
Intension and strong inheritance: corporate sub-type inherits the identification features of type Reservation (intension).
Mind maps built on these principles could provide a common thesaurus encompassing the whole of enterprise data, information and knowledge.
Intelligence: Data, Information, Knowledge
Considering that mind maps combine intelligence and cartography, they may have some use for enterprise architects, in particular with regard to economic intelligence, i.e the integration of information processing, from data mining to knowledge management and decision-making:
Data provide the raw input, without clear structures or semantics (terms or aspects).
Categories are used to process data into information on one hand (extensional nodes), design production systems on the other hand (intensional nodes).
Abstractions (concepts) makes knowledge from information by putting it to use.
Along that perspective mind maps could serve as front-ends for enterprise architecture ontologies, offering a layered cartography that could be organized according to concerns:
Enterprise architects would look at physical environments, business processes, and functional and technical systems architectures.
Knowledge managers would take a different perspective and organize the maps according to the nature and source of data, information, and knowledge.intelligence w
As demonstrated by geographic information systems, maps built on clear semantics can be combined to serve a wide range of purposes; furthering the analogy with geolocation assistants, layered mind maps could be annotated with punctuation marks (e.g ?, !, …) in order to support problem-solving and decision-making.
Data mining explores business opportunities and competitive advantage, requirements analysis considers supporting applications. Both use models, the former’s are predictive and ephemeral, the latter’s descriptive (or prescriptive) and perennial.
As the generalization of digitized environment calls for more integration of business and software engineering processes, understanding the relationship between data mining and requirements analysis could significantly improve processes maturity and agility.
Data vs Requirements Analysis
Nowadays the success of a wide range of enterprises critically depends on two achievements:
Mapping business models to changing environments by sorting through facts, capturing the relevant data, and processing the whole into meaningful and up to date information. That can be achieved through analysis models mapping business expectations to supporting systems.
Putting that information into effective use through business processes and supporting systems. That is done through systems architecture and design models meant to prescribe how to build software artifacts.
Those challenges are converging: under the pressure of markets forces and technological advances most of traditional fences between business channels and IT systems are crumbling, putting the focus on the functional integration between data mining and production systems. That’s where predictive models can help by anchoring descriptive models to moving markets and by cross-feeding analysis and operations. How that can be achieved has been the bread and butter of good corporate governance for some time, but there has been less interest for the third branch, namely how data analysis (predictive models) could “inform” business requirements (descriptive models).
Facts are not given but must be captured through a symbolic description of actual observations. That entails some observer set on task using a mix of conceptual and technical apparatus. Data mining and requirements analysis are practical realizations of that process:
Data mining relies on analytic tools to extract revealing information that could be used to chart opportunities along business models.
Requirements analysis relies on business processes and users’ practice to extract symbolic descriptions that will be used to build models of supporting applications.
If both walk the path from data to information, their objectives are different: the former’s is to improve business decisions by making sense of actual observations; the latter’s is to build system surrogates from the symbolic descriptions of actual business objects and activities.
Anchors & Structures: Plasticity of Business Entities
Perhaps paradoxically, business agility calls for terra firma because nimble trades must be rooted in corporate identity and business continuity. As a consequence, the first step of requirements analysis should be to associate individuals business objects or activities with stable and consistent identification mechanisms, and to group them with regard to that mechanism:
External entities with natural (person) or designed identity (car).
Symbolic entities for roles (customer) or commitments (maintenance contract).
Actual activities (promotion campaign) and events (sale) or business logic (promotion).
Conversely, as the aim of data analysis is to explore every business angle, individual observations are supposed to be moved across groups; yet, since the units identified by data analysis will have to be aligned with the ones described by requirements analysis, moves must also keep track of identities. That dilemma between continuity of identified structures on one side, plasticity of functional aspects on the other side, can be illustrated by banks which, in response to marketing requirements, had to shift from account (internal identification) to customer (external identification) based systems.
That challenge can be overcome by linking the identification of symbolic entities to external anchors.
Profiles & Features: Versatility of Business Opportunities
As noted above, requirements and data analysis are set on the same road but driven by different forces: the former tries to group individuals with regard to identification mechanisms before fleshing them out with relevant features; the latter tries to group individuals with given identities according to features and opportunity profiles. Yet, what could appear as collision courses may become a meeting of minds if both courses are charted with regard to variants analysis.
From the requirements perspective the primary concern is to distinguish between structural and functional variants:
Structural variants are bound to identities, i.e set up-front for the respective life-cycle of individual business objects or transactions. As a consequence they cannot be changed without undermining business continuity. Moreover, being part and parcel of descriptors (e.g types and use cases) their change will affect engineering processes.
Functional variants may vary during the respective life-cycle of individual business objects or transactions. As a consequence they can be changed without undermining business continuity, and changes in descriptors (e.g partitions and scenarii) can be managed without affecting engineering processes.
From the data mining perspective the objective is to improve the benefits of information systems for decision-making processes:
Static: how to classify individuals as to reduce the uncertainty of predictions
Dynamic: how to classify business options as to reduce the uncertainty of decisions.
Since those objectives are set for individuals, constraints on continuity and consistency can be dealt with independently of the description of symbolic surrogates.
It ensues that perspectives can be adjusted by factoring out the constraints of continuity and consistency for business objects (e.g cars), agents (e.g customer) and processes (e.g repair). Profiles for agents (a), behaviors (b), and business options (c) could then be freely explored and tailored with regard to changes in business environment and objectives.
Applying Data Analysis to Requirements
Not surprisingly data analysis techniques can be used to adjust perspectives. For that purpose a sample of individuals (business objects and operations) representing the population targeted by requirements would have to be submitted to basic mining routines. Borrowing a catalog from F. Provost & T. Fawcett:
Classification: estimates the probability for each individual (objects or operations) to belong to a set of classes; can be used to assess the closeness of the variants (respectively power-types or execution paths) identified by requirements analysis.
Regression: reverse classification; estimates how much of individual features valuations can be explained by the proposed classifications.
Similarity: a shallow version of classification; can be used to assess the distance between variants and consolidate the proposed classifications.
Clustering: a deep version of classification; can be used to distinguish between shallow and natural classifications.
Co-occurrence: deals with behavioral variants; can be used to distinguish between functional and structural classifications.
Profiling: reverse of co-occurrence; can be used to consolidate functional and structural classifications.
Links prediction: can be used to define relationships.
Data reduction: eliminate redundant individuals; can be used to consolidate requirements and refine tests scenarii.
Causal modeling: brings together business logic (events and rules) and users decisions; should provide the backbone of tests scenarii.
Besides the direct benefits for requirements, such procedures may help to bridge the span between data and requirements analysis and significantly improve processes’ capability and maturity level.
Data mining being first and foremost about competitive edge, it relies on a timely and effective coupling between enterprises capabilities and business opportunities. But the dilemma between continuity and plasticity described above for business objects and processes reappears at enterprise level: how to conciliate architecture, by nature perennial, with the agility needed to make the best of changing and competitive environments ?
As architectural big bang is arguably a last resort option, answers to that question must be progressive and local: if changes are to be swift and pertinent they must be both circumscribed and leveraged to the relevant parts of architecture. Taking an (amended) leaf of the Zachman framework, its sixth column (“Why” ) could be reset as a line for business and operational objectives that would cross the original five columns instead of the architecture layers. Using a pentagonal representation of enterprise architecture, that line would be set as circling the outer range.
It is worth to note that setting objectives on a line crossing the columns of capabilities instead of a column crossing the lines of layers means that objectives are set at enterprise level and their cascading impact traced and managed through layers.
Conceptual Models & Business Contexts
But even that updated framework doesn’t take into account the fundamental changes in business environments. Once secure behind organizational and technical fences, enterprises must now navigate through open digitized business environments and markets. For business processes it means a seamless integration with supporting applications; for corporate governance it means keeping track of heterogeneous and changing business contexts and concerns while assessing the capability of organizations and systems to cope, adjust, and improve.
As long as environments were a hotchpotch of actual and symbolic artifacts the pros and cons of integration could be balanced. But the generalization of digital flows and transactions has upended the balance: there is no more room or time for latency and enterprises must bring all symbolic representations (business, organization, and systems) under a common conceptual roof:
A canonical approach would be to introduce a conceptual indexing scheme open to extensions but with its footprint defined by business processes and systems functionalities. That would ensure a better integration of processes and supporting engineering but will do nothing for the modeling gap between enterprise architecture and external contexts. That could be achieved with ontologies.
Conceptual Loop: Ontologies & Business Intelligence
As far as data mining is concerned, three kinds of operations are to be considered:
Data understanding gives form and semantics to raw material.
Business understanding charts business contexts and concerns in terms of objects and processes descriptions.
Modeling consolidate data and business understanding into descriptive, predictive, and operational models.
While traditional approaches fall short when tasked with iterative modeling of unstructured data, ontologies may fare better because their explicit aim is only to describe what could exist in a domain of discourse:
They are made of categories of things, beings, or phenomena; as such they may range from simple catalogs to philosophical doctrines.
They are driven by cognitive (i.e non empirical) purposes, namely the validity and consistency of symbolic representations.
They are meant to be directed at specific domains of concerns, whatever they can be: politics, religion, business, astrology, etc.
With regard to models, only the second point puts ontologies apart: contrary to models, ontologies are about understanding and are not supposed to be driven by empirical purposes. It ensues that ontologies can be understood as conceptual (aka canonical) models, used as such for business analysis and extended with purposes for systems analysis and design.
In addition to the integration with enterprise architectures, ontologies benefits for business intelligence would be twofold.
On one side, and whatever their use, ontologies could be aligned with the nature of contexts and their impact on business and enterprise governance e.g:
Institutional: mandatory semantics sanctioned by regulatory authority, steady, changes subject to established procedures.
Professional: agreed upon semantics between parties, steady, changes subject to established procedures.
Corporate: enterprise defined semantics, changes subject to internal decision-making.
Social: definedpragmatic semantics, no authority, volatile, continuous and informal changes.
Personal: customary semantics defined by named individuals.
On the other side ontologies could be defined according to the nature of targeted items, namely terms, documents, symbolic representations, or actual objects and phenomena. That would outline four basic concerns that may or may not be combined:
Thesaurus: ontologies covering terms and concepts.
Document Management: ontologies covering documents with regard to topics.
Organization and Business: ontologies pertaining to enterprise organization, objects and activities.
Engineering: ontologies pertaining to the symbolic representation of products and services.
On a broader perspective, ontologies could be influential in spanning the gap between explicit and implicit knowledge. In a few years’ time practically unlimited access to raw data and the exponential growth in computing power have opened the door to massive sources of unexplored knowledge which is paradoxically both directly relevant yet devoid of immediate meaning:
Relevance: mined raw data is supposed to reflect the geology and dynamics of targeted markets.
Meaning: the main value of that knowledge rests on its implicit nature; applying existing semantics would add little to existing knowledge.
Assuming that deep learning can transmute raw base metals into knowledge gold, ontologies would be decisive in framing the understanding, assessment, and improvement of the processes.
Operational Loop: Business Intelligence & Decision-making
Once carried out separately and periodically, decision-making is to be carried out iteratively at operational, tactical, and strategic level; while each level is to be set along its own time-frames, all are to rely on data-mining, with cycles following the same pattern:
Observation: understanding of changes in business opportunities.
Orientation: assessment of the reliability and shelf-life of pertaining information with regard to current positions and operations.
Decision: weighting of options with regard to enterprise capabilities and broader objectives.
Action: carrying out of decisions within the relevant time-frame.
Given business new playground, decision-making processes have to weave together material and digitized flows, actual contexts (aka territories) and symbolic descriptions (maps), and overlapping time-frames (operational tactical, strategic). That operational loop could then be coupled with the broader one of business intelligence: