Table 3 Comparing the conceptual model... | Emerald Publishing

Table 3

Comparing the conceptual model with the empirical data

SLT	R		C		In		Description of the comparison between the conceptual model (CM) and the empirical data (ED)
SLT	CM	ED	CM	ED	CM	ED
$E_{k}$			✓	✓	✓	✓	The ED support Wikner's (2015, 2018) work, suggesting that the length of $E$ has implications for a company's investments and inventory-carrying costs of raw materials, for instance. However, the ED stress the importance of considering not only the length but also uncertainties in $E$ ⁠, as these have significant implications for investment and cost related to safety buffers, in line with the findings of Blackburn (2012) and Er and MacCarthy (2006)
$I_{k}$			✓	✓	✓	✓	In terms of $I$ ⁠, the findings suggest that the length of and uncertainty in $I$ will have implications for cost and investment in terms of capacity. For instance, the respondents argued that in many cases, a longer $I$ was a result of larger lot sizes in order to decrease set-up costs and use capacity more efficiently, in which more materials (i.e. WIP) accumulated within the system, increasing both the inventory-carrying costs and the investments made in WIP. This is related to Stalk and Hout's (1990) 0.05–5% rule, also supporting Wouters' (1991) results
$S$				✓	✓	✓	The upstream end of $S$ indicates the point in the flow at which the company must start taking a material-based risk. $S$ ⁠, especially the $S - D$ segment, is therefore related to the amount of investment in the materials within the system (Christensen et al., 2007; Wikner, 2015). This is supported by the empirical findings, in which the respondents argued that a longer $S$ would usually result in more materials accumulating in the system. However, they also recognised the cost of investing in these materials. They added that a consequence of a longer $S$ would be a system that would be less responsive to product and technology changes, exacerbating costs further, especially for companies with shorter product life cycles. The empirical findings thus suggest that $S$ has strong implications for both cost and investment, rather than just investment, as discussed by Wikner (2015)
$D$	✓	✓	✓	✓	✓	✓	For its part, $D$ determines the position of the CODP and of the related inventory point and consequently, the capital tied up in the materials (Hedenstierna and Ng, 2011; Hoekstra and Romme, 1992). Furthermore, an improved accuracy in $D$ (improved certainty in lead time, not volume) would reduce the needed safety buffer at the CODP, consequently reducing the inventory-carrying cost (Forza et al., 2008; Gregory and Rawling, 1997; Mather, 1988; Wouters, 1991). This is supported by the respondents, who recognised both the investment and the cost in the CODP buffer and/or the capacity dimensioning downstream of the CODP. Moreover, the CM states that $D$ represents a key lead time for creating a competitive advantage (Hedenstierna and Ng, 2011; Hoekstra and Romme, 1992; Vickery et al., 1995), which is an opportunity for boosting revenue by increasing the sales volume and/or the selling prices (Wouters, 1991). The respondents agreed, adding that this implication might be stronger if $D$ was an order winner in terms of length or certainty, for example. All respondents discussed $D$ in relation to the other SLTs and called for reducing these SLTs and the decision points downstream of $D$ (downstream of the CODP) because of the costs associated with the risk of performing activities on speculation. Thus, in a sense, they emphasised the importance of $D$ as an SLT
$A_{S, i}$	✓	✓	✓	✓		✓	Agreeing with the CM (Gregory and Rawling, 1997; Wikner, 2015), the respondents said that reducing an $A_{S}$ such that it would be pushed downstream of the CODP, becoming an $A_{D}$ candidate, could potentially increase revenue. R1 explained that a revenue increase could stem from a higher demand and the ability to charge a premium price. However, the implications of $A_{S}$ for revenue could be regarded as indirect because once an $A_{S}$ becomes an $A_{D}$ ⁠, it is $A_{D}$ that has implications for revenue. However, the respondents also indicated that this untapped possibility for increasing revenue might not be realised easily if the customisations had not been requested by the market. In such a case, the potential cost implications would probably be more significant, as these would result in more item numbers, for instance. The respondents also mentioned that costs could increase, owing to the expenses incurred in setting up and maintaining new items. However, deviating from Wikner (2015), the respondents also argued that higher investments would be needed to stock up on the additional items
$A_{D}$	✓	✓		✓		✓	One of the main differences between the CM and the empirical findings is related to $A_{D}$ ⁠. Gregory and Rawling (1997) and Wikner (2015) argued that $A_{D}$ had an obvious potential to generate more revenue, a view shared by the respondents. However, R8 from AircraftCo did not perceive a relation between the length of $A_{D}$ and a company's revenue-generating capability. R8 believed that offering customisation within the requested $D$ would have implications for revenue, but the length of $A_{D}$ per se would have none. Nonetheless, Er and MacCarthy (2006), Forza et al. (2008) and Wikner (2015) suggested that if the point of differentiation could be moved downstream (i.e. if $A_{D}$ was reduced), then the customer’s input for product differentiation would become necessary at a later point. Thus, the customer can place an order at an earlier point but decide on the customisation specifics later, increasing the order flexibility and the possible revenue. However, the respondents presented an aspect not covered in the CM. Specifically, if a reduced $A_{D}$ could be obtained through postponement (see, e.g. Forza et al., 2008), for example, then this could result in reduced investments and inventory-carrying costs, in which materials and components could be acquired and/or produced on commitment to customer orders instead of being maintained in the inventory

SLT	R		C		In		Description of the comparison between the conceptual model (CM) and the empirical data (ED)
SLT	CM	ED	CM	ED	CM	ED
$E_{k}$			✓	✓	✓	✓	The ED support Wikner's (2015, 2018) work, suggesting that the length of $E$ has implications for a company's investments and inventory-carrying costs of raw materials, for instance. However, the ED stress the importance of considering not only the length but also uncertainties in $E$ , as these have significant implications for investment and cost related to safety buffers, in line with the findings of Blackburn (2012) and Er and MacCarthy (2006)
$I_{k}$			✓	✓	✓	✓	In terms of $I$ , the findings suggest that the length of and uncertainty in $I$ will have implications for cost and investment in terms of capacity. For instance, the respondents argued that in many cases, a longer $I$ was a result of larger lot sizes in order to decrease set-up costs and use capacity more efficiently, in which more materials (i.e. WIP) accumulated within the system, increasing both the inventory-carrying costs and the investments made in WIP. This is related to Stalk and Hout's (1990) 0.05–5% rule, also supporting Wouters' (1991) results
$S$				✓	✓	✓	The upstream end of $S$ indicates the point in the flow at which the company must start taking a material-based risk. $S$ , especially the $S - D$ segment, is therefore related to the amount of investment in the materials within the system (Christensen et al., 2007; Wikner, 2015). This is supported by the empirical findings, in which the respondents argued that a longer $S$ would usually result in more materials accumulating in the system. However, they also recognised the cost of investing in these materials. They added that a consequence of a longer $S$ would be a system that would be less responsive to product and technology changes, exacerbating costs further, especially for companies with shorter product life cycles. The empirical findings thus suggest that $S$ has strong implications for both cost and investment, rather than just investment, as discussed by Wikner (2015)
$D$	✓	✓	✓	✓	✓	✓	For its part, $D$ determines the position of the CODP and of the related inventory point and consequently, the capital tied up in the materials (Hedenstierna and Ng, 2011; Hoekstra and Romme, 1992). Furthermore, an improved accuracy in $D$ (improved certainty in lead time, not volume) would reduce the needed safety buffer at the CODP, consequently reducing the inventory-carrying cost (Forza et al., 2008; Gregory and Rawling, 1997; Mather, 1988; Wouters, 1991). This is supported by the respondents, who recognised both the investment and the cost in the CODP buffer and/or the capacity dimensioning downstream of the CODP. Moreover, the CM states that $D$ represents a key lead time for creating a competitive advantage (Hedenstierna and Ng, 2011; Hoekstra and Romme, 1992; Vickery et al., 1995), which is an opportunity for boosting revenue by increasing the sales volume and/or the selling prices (Wouters, 1991). The respondents agreed, adding that this implication might be stronger if $D$ was an order winner in terms of length or certainty, for example. All respondents discussed $D$ in relation to the other SLTs and called for reducing these SLTs and the decision points downstream of $D$ (downstream of the CODP) because of the costs associated with the risk of performing activities on speculation. Thus, in a sense, they emphasised the importance of $D$ as an SLT
$A_{S, i}$	✓	✓	✓	✓		✓	Agreeing with the CM (Gregory and Rawling, 1997; Wikner, 2015), the respondents said that reducing an $A_{S}$ such that it would be pushed downstream of the CODP, becoming an $A_{D}$ candidate, could potentially increase revenue. R1 explained that a revenue increase could stem from a higher demand and the ability to charge a premium price. However, the implications of $A_{S}$ for revenue could be regarded as indirect because once an $A_{S}$ becomes an $A_{D}$ , it is $A_{D}$ that has implications for revenue. However, the respondents also indicated that this untapped possibility for increasing revenue might not be realised easily if the customisations had not been requested by the market. In such a case, the potential cost implications would probably be more significant, as these would result in more item numbers, for instance. The respondents also mentioned that costs could increase, owing to the expenses incurred in setting up and maintaining new items. However, deviating from Wikner (2015), the respondents also argued that higher investments would be needed to stock up on the additional items
$A_{D}$	✓	✓		✓		✓	One of the main differences between the CM and the empirical findings is related to $A_{D}$ . Gregory and Rawling (1997) and Wikner (2015) argued that $A_{D}$ had an obvious potential to generate more revenue, a view shared by the respondents. However, R8 from AircraftCo did not perceive a relation between the length of $A_{D}$ and a company's revenue-generating capability. R8 believed that offering customisation within the requested $D$ would have implications for revenue, but the length of $A_{D}$ per se would have none. Nonetheless, Er and MacCarthy (2006), Forza et al. (2008) and Wikner (2015) suggested that if the point of differentiation could be moved downstream (i.e. if $A_{D}$ was reduced), then the customer’s input for product differentiation would become necessary at a later point. Thus, the customer can place an order at an earlier point but decide on the customisation specifics later, increasing the order flexibility and the possible revenue. However, the respondents presented an aspect not covered in the CM. Specifically, if a reduced $A_{D}$ could be obtained through postponement (see, e.g. Forza et al., 2008), for example, then this could result in reduced investments and inventory-carrying costs, in which materials and components could be acquired and/or produced on commitment to customer orders instead of being maintained in the inventory

Note(s): C = cost, In. = investment, R = revenue