CBP BPA cheat sheet

BPMN 2.0 core elements: Events (start/intermediate/end), Activities (task/subprocess), Gateways (exclusive, inclusive, parallel, event-based), Sequence Flows, Message Flows, Pools/Lanes.

Exclusive gateway (XOR) routes exactly one outgoing path based on conditions; inclusive gateway (OR) can activate one or more paths; parallel gateway (AND) activates all paths and waits for all incoming tokens.

A process model is a set of activities, events, gateways, and control-flow connections that represent the order and conditions of work; it is not a data model or organizational chart.

Token simulation: a token represents a case at a point in the process; gateways split/merge tokens; AND-split creates a token on every outgoing flow, AND-join waits for a token on every incoming flow.

Modeling levels: descriptive (business view), analytical (enough for simulation/analysis), executable (detailed enough for engine execution); choose level based on purpose.

BPMN pools represent participants; lanes represent roles or departments within a pool; message flows cross pool boundaries; sequence flows cannot cross pool boundaries.

Event types: start, intermediate (catch/throw), end; triggers include message, timer, error, signal, escalation, compensation, conditional, link.

Process discovery techniques: interviews, observation, document analysis, system log extraction (process mining); output is an as-is model validated with stakeholders.

Model quality criteria: syntactic correctness, semantic correctness, completeness, consistency, and understandability; use a modeling guideline to enforce naming and layout conventions.

Little's Law: WIP = Throughput × Cycle Time (average number of items in process equals arrival rate times average time in system).

Cycle time = 1 / throughput (for a stable process); throughput is the rate at which completed items exit the process.

Bottleneck is the resource or activity with the highest utilization or longest cycle time; process capacity is limited by the bottleneck.

Utilization = (arrival rate × processing time) / number of resources; utilization near 100% causes long queues (queueing theory).

Value-added analysis: classify activities as value-adding, business-value-adding (non-value-adding but required), or non-value-adding (waste).

Eight wastes (TIMWOODS): Transportation, Inventory, Motion, Waiting, Overproduction, Overprocessing, Defects, Skills underutilization.

Process mining metrics: conformance (does log match model), performance (bottlenecks, waiting times), and organizational (resource interactions).

Root cause tools: 5 Whys, Fishbone (Ishikawa) diagram, Pareto analysis (80/20), FMEA (Failure Mode and Effects Analysis).

Simulation inputs: arrival distribution, processing times, resource counts, routing probabilities; validate model against observed cycle time and throughput.

Design principles: eliminate handoffs, reduce wait states, automate repetitive tasks, standardize exceptions, and design for the customer's moment of truth.

To-be process modeling: start from as-is analysis, apply redesign heuristics (e.g., eliminate, simplify, combine, automate), then validate with stakeholders.

Business rules and decision logic: use decision tables or DMN (Decision Model and Notation) to separate rules from process flow.

Service design blueprinting: front-stage (customer-facing) and back-stage (support) actions, with line of interaction, line of visibility, and line of internal interaction.

Process ownership and governance: define process owner, KPIs, SLAs, and escalation paths; align with organizational roles and RACI.

Risk and control design: identify control points, segregation of duties, approval thresholds, and audit trails; embed controls into the process model.

Human vs. automated task allocation: consider volume, variability, judgment, and cost; use RPA for rule-based, high-volume, stable tasks.

Design for exception handling: model error events, compensation, and escalation paths; avoid designing only the happy path.

Transformation levers: process redesign, technology enablement (RPA, BPMS, AI), organizational change, and data/analytics.

Implementation roadmap: prioritize initiatives by impact and feasibility (e.g., impact/effort matrix), then sequence quick wins and strategic projects.

Change management models: Kotter's 8 steps, ADKAR (Awareness, Desire, Knowledge, Ability, Reinforcement), and Lewin's unfreeze-change-refreeze.

RPA suitability: rule-based, repetitive, high-volume, stable, digital-input tasks; avoid tasks requiring judgment or unstructured data.

BPMS (Business Process Management Suite) supports modeling, execution, monitoring, and optimization; process mining can feed continuous improvement.

KPIs for transformation: cycle time reduction, cost per transaction, error rate, customer satisfaction (NPS/CSAT), and adoption rate.

Governance for transformation: executive sponsorship, PMO/CoE (Center of Excellence), stage gates, and benefits realization tracking.

Risk of transformation: scope creep, resistance to change, technology fit, data quality, and process drift after go-live.