1 Intended And Actual Variability In The Product Portfolio
This section reproduces portions of chapter 2 of VARIES deliverable D3.5 (Van den Broeke, et al., 2015), and has been adapted for public dissemination.
Intended variability in a product portfolio is the variability that was intended by design at a given moment in the past, typically for a point in time in the future. Actual variability reflects the reality at a given point in time. This is illustrated in Figure 1‑1, in which products in the portfolio are represented by shapes.
Referring to the example from Figure 1‑1, at a given point in time tA the (actual) product portfolio at time tA consists of product variants (a), (b) and (c) (coloured dark blue). On a portfolio planning exercise for time tB in the future, the intention was formulated to phase out product variants (a) and (b) (coloured blank), to keep (c) and to create a new product variant (d) (coloured light blue). Eventually, at time tB (after tA) the content of the product portfolio is assessed and it appears that product variant (b) has not yet been phased out (coloured orange), product variants (c) and (d) (coloured green) are available as intended but a new product variant (e) (coloured orange) also made it in the current product portfolio at time tB. It is clear that in this case the product portfolio intended for time tB differs from the actual product portfolio at time tB.
This approach can of course be chained starting from tB and ending in tC etc.
For completeness, the possible discrepancies between intended and actual product portfolio can be found in Table 1‑1:
|Decision at tA for tB||Reality at tB||Description|
|Create product||Created||(As predicted)|
|Not created||This product variant was not created|
|Preserve product||Preserved||(As predicted)|
|Removed||This product variant was not preserved|
|Remove product||Removed||(As predicted)|
|Not removed||This product variant was not removed|
|(no decision)||Created||This product variant was not anticipated at tA (or it may have been planned at a later point in time tC)|
The main reasons for differences between intended and actual product portfolio are linked to the inherent uncertainty in predicting the future or in correctly assessing the impact of portfolio decisions on future benefits, while keeping the variability paradox under control.
How can the benefits offered by introducing variability into embedded systems
outweigh the increased product complexity caused by variability
Other reasons can be that it was not possible to perform an in-depth assessment of the portfolio in a roadmapping exercise, or that the assessment was done after the portfolio definition exercise. Sometimes the assessment has only be performed partly due to various constraints.
The portfolio decisions presented in Table 1‑1 result in the definition of variation points and (future) product variants. It will be important to evaluate the cost of adding more complexity by adding variation points (that may eventually never be realised) with respect to the benefit of facilitating the creation of (predictable and unpredictable) product variants.
2 Product Portfolio Scenarios
Figure 2‑1 shows a situation in which two product portfolio scenarios are proposed at time tB: from product portfolio A to B1 or from product portfolio A to B2.
3 Product Roadmapping Scenarios
Starting from a current product portfolio A there are several possible paths to establish an intended future product portfolio B, as illustrated in Figure 3‑1.
Along these decision paths are a set of variability decisions (e.g. add a product). Although many paths may exist, only a few paths will be of relevance due to economic, technical, time and other constraints. The example of Figure 3‑1 illustrates a situation in which two variability decision paths, which have been highlighted with dotted arrows, are considered for further evaluation. Imagine that a company performed an analysis of likely market evolutions and their impact on the way the product portfolio should evolve over time. Decision path 1 could for instance prioritize a shorter time to market for a number of features, where decision path 2 attempts at minimizing the engineering costs.
The variability decision sequence (path) from Figure 3‑1 will result in a planning of product additions, modifications and eliminations over time. This is depicted in Figure 3‑2, which represents the planning of intended product releases and variant releases in the future. Products are represented by shapes, variants of a same product are represented by a prime (e.g. b → b’). Times at which a product will be added (e.g., tA1), removed (e.g. tA2) or replaced (e.g., tA3) are indicated on the timeline along with the symbol representing the product affected.