In a business operational environment, there are three reasons for work:
- To deliver goods or services - Operational State
- To assure the availability of resources - Operational State
- To change the nature of the operational environment - Programs/Projects
The TEM Modeler facilitates the detailed capture of the first two and provides transforms to investigate future alternatives. The Modeler also provides financially abstracted views (scenarios) to capture the cost and time essence of programs/projects.
The following discussion is about work. A generally accepted definition of work is:
activity in which strength or faculties are applied to do or perform something.
This short description combines three things into one: an objective, energy, and resources. The definition does not imply granularity (the concept of an atomic-level of work). It does, however, imply that resource usage time is intrinsic. The verbs do/perform imply that the work has a finish time relative to a start time. The resource(s) are applied in various portions during this time period. In the TEM Modeler, a resource is defined as any reusable object that facilitates work. Since all resources have inherent cost, time of usage allows us to measure and compare the cost of work.
Work instances are often called activities, tasks, work units, etc., to denote atomic units of delivery. Note that some interpretations of the word activities implies they can contain subtasks and are, therefore, not atomic in nature. We will predominantly use the term work unit which has atomic granularity.
Work units have frequency and duration.
The work unit's frequency is always proportional to the work driver's rate or count. Hence, changes in the driver count or rate cause proportional changes in the work unit's frequency:
where f1 is the work unit's original frequency, dr2 is the new driver count, dr1 is the original driver count, and f2 is the new frequency.
The work unit's duration is exponentially related to human maturity. TEM assumes a maturity range of 1 <= m <=5 and uses the formula
where C is an empirically derived constant, m1 is the old maturity, m2 is the new maturity, and d1 is the old duration.
The TEM Modeler has a Monte Carlo simulation engine that lets one specify two probability distribution functions: one for the work unit's frequency and one for the work unit's duration. Hence, one is able to investigate resource contention issues.
Cost/Revenue templates are associated with work units and determine how consumables and revenue are accrued. Accruals are classified as Fixed-Annual, Instance-Based, Duration-Based, or Duration Plus Delay-Based.
Work units have a when offered time represented as a triple (hours per day, days per week, weeks per year).
Each resource is utilized for some period of time to deliver the work unit. This resource utilization is specified as a positive real number. This provides for the cases where:
- a work unit requires more than one resource from the same resource group, and,
- a resource is required for a fraction of the work unit’s duration.
Resources have a when available time defined in terms of a triple (hours per day, days per week, weeks per year). This triple also defines the resource’s capacity. Hence, in the USA, employees will typically have an availability period of (8, 5, 50). Similarly, a 3D printing machine might have an availability of (24, 7, 52). The Modeler lets the user apply multiple resources to any work unit, resolves the when offered and when available, and informs the user when the minimal resources are not available.
Resources have a cost rate that is used when calculating the utilized cost. It is a rate because the resource cost is specified in currency per unit of time. If one sums the resource time usage across all work units where that resource is applied, one can calculate the resource utilization cost by
where UR is the utilization of resource R, is the resource usage for that work unit, and Rcapacity is the resource capacity in terms of its when available.
Resources have a classification represented by a driver. As a result, one can accrue the amount of work necessary to assure the availability of those resources.
All work is performed as a response to a driver. A driver has two manifestations: a rate representing flow and a count representing the number of similar resources.
Several years ago, we were in meetings whereby multiple Information Technology (IT) projects were being assessed and compared. As was typical at that time, each operational project was characterized with strategic, tactical, and, financial impact metrics. The first two were very much experienced-based metrics where the senior managers brought years of insight to bear. The last metric, however, was based on the current spend, project cost, and the future spend.
Unfortunately, future spend was, more often than not, presented in terms of productivity improvement. Automation, process improvements, outsourcing were stated in terms of cost savings related to labor reduction. Very little work was done to show the mechanics of either the current spend or the future spend.
What was needed was a way to easily describe the current state in terms of organizations, resources and work and then relate it to one or more future states. But how should we represent what is accomplished in the current and possible multiple future alternatives? And how should this representation be cost tied to the various organizations and to the various delivery processes?
Our (eventual) solution was to capture the characteristics of the typical operational environment in order to:
- Fully describe work and the role resources, consumables, and revenue play
- Depict how the accomplishment of work requires at least one and, more often than not, multiple resources
- Show how work is driven by resource count and throughput rates
- Show how the work relates to organizations (bottom up) and to catalogs (top down)
- Describe how resources are sourced to one ...or more organization(s)
- Explain how cost of work accrued to an organizational roll up should match exactly the cost of work accrued to products and services in the catalog
- Show how organizations may be abstracted into providers that afford an opportunity for cost localization and catalog ownership
One should recognize the above as an extended, rate-based ABC (activity based costing) set of properties.
The implemented solution uses well known technology and programming techniques:
- A reasonably simple database schema that is implemented in two technologies
- A spreadsheet-like analysis engine to quickly analyze resource usage
- A Monte-Carlo simulation engine to better understand resource contention and usage variance
- Two hierarchical views to facilitate top down or bottom up data entry and analysis
- A generalized linear transformation mechanism as an adjunct to organizational copy and paste
- A cost analysis engine that traverses hierarchy and maps utilized cost to actual costs
The resulting software models the operational environment using familiar objects. The two views, top-down and bottom-up, network-wise and hierarchically cover the work performed and, therefore, provide cost credibility by each view cost summing to the same value. This is explained in the image below.