探索供應鏈架構下企業機會之分析：從採購與生產之角度

(1)

科技部補助專題研究計畫成果報告

期末報告

探索供應鏈架構下企業機會之分析：從採購與生產之角度

計畫類別：個別型計畫計畫編號： MOST 103-2410-H-004-107-執行期間： 103年08月01日至104年12月31日執行單位：國立政治大學企業管理學系計畫主持人：陳立民計畫參與人員：碩士班研究生-兼任助理人員：呂杰大專生-兼任助理人員：許嘉宏報告附件：出席國際會議研究心得報告及發表論文處理方式： 1.公開資訊：本計畫可公開查詢 2.「本研究」是否已有嚴重損及公共利益之發現：否 3.「本報告」是否建議提供政府單位施政參考：否

中　華　民　國　105　年　02　月　02　日

(2)

中文摘要：本研究結合存貨理論與創業精神來延伸研究,藉由此我們發表兩篇文章與兩篇上獲得修改的機會

中文關鍵詞：存貨管理,創業精神

英文摘要： This research combines inventory management and

entrepreneurship. With the funding support, we publish two papers and get two other manuscripts for revision.

(3)

探索供應鏈架構下企業機會之分析:從採購與生

產的角度

(4)

With the funding support, we are very glad that two papers have been published in 2015 and two other papers received the revision opportunity.

Publication

The first one is the work “Robust Supply Chain Strategies for Recovering from

Unanticipated Disasters” published in Transportation Research Part E: Logistics and Transportation Review (vol 77, Issue C, page 198-214, 2015). This journal is

listed as a recommendation journal from Ministry of Science and technology with impact factor: 2.676. In this work, we try to understand a critical issue in today’s business: how do supply chain members recover from unanticipated disaster? This paper examines the effectiveness of popular recovery strategies used to address disasters in a supply chain. We create a cellular automata model to portray the dynamic operational performance among supply chain members facing disruptions caused by natural and man-made disasters. Our analysis shows that a supply chain achieves the best recovery performance if members adopt a radical strategy that fixed the disruption immediately. This observation is robust to various resource consumption requirements. Our findings are translated into a managerial framework for supply chain managers by formulating strategies for recovering from unanticipated disasters

We also have another work “Process Innovation and Improvement: A

Competitive Dynamics Perspective” recently accepted by NTU Management

Review (台大管理論叢) in January. 2015 (expected to publish in 2017 informed by the editor). The main contribution of this work is as follows. Operations and strategy literatures have consistently promoted the best practice of accredited management standards for process efficiency and effectiveness. Based on the capability theorizing, studies have investigated how a following firm can improve its operational performance by learning from a leading firm’s best practices. Our study extends this research stream by applying competitive dynamics perspective from strategy literature to the context of process development and management. We develop insights by applying a dynamic, computational model based on an extensive appraisal of the history of process innovation and improvement in the global automobile industry and draws on the underlying theoretical relationships in the empirical literatures on operations and strategy. The core proposition in our study is that a follower firm’s investment in process innovation capabilities for long-term growth will provoke strong retaliation from leading firms using the current best practice. We demonstrate that the leading firm can publicly signal its best practice to induce the follower firm to invest in process improvement capabilities but only for short-term survival, not for long-term purposes or goals. In this way, the leader firm maintains its leading edge.

(5)

Our results also underscore the importance of competition in determining the firm-level process development and management practices.

Work Under Revision

In addition, we have two works under revision and both are submitted to top journals in operations management domain.

The paper “Is Cloud Bursting (a Hybrid Cloud) Advantageous to a Service Firm with On-site Capacity?” has received the revision opportunity in July, 2015 by

Decision Sciences Institute (DS). With two months effort, we re-submit it in

September 2015. This work is motivated by the popular adoption of cloud bursting, a hybrid computing resource model in which a firm first uses its own internal computing resources and then bursts to a public cloud for extra resources as needed. Thus, this paper develops quantitative models in non-competitive and competitive environments to examine whether a service firm that has on-site capacity to support its online service (e.g., applications and online shopping) to end clients/customers can benefit from cloud bursting. If a firm adopts a hybrid cloud, it benefits from immediate, extra computing resources, and scalability, but it assumes potential risks (i.e., security threats or/and outage) from this external resource. If a firm uses only a private cloud, while it has no risk concerns, it may miss the sales opportunities at peak service times due to limited resource capacity. Overall, we show that a profit-maximizing firm prefers a hybrid cloud if risk is considerably low while it would benefit from a private cloud if risk is considerably high. Interestingly, a firm under competition may counterintuitvely deploy a private cloud even though risk remains low while its competitor still deploys a hybrid cloud. Our numerical study conducts the sensitivity analysis to link a firm’s profit performance with its best cloud deployment in both non-competition and competition environments.

In addition, we just have another work “Are Purchasing-trigger Donations Advantageous with Competition?” acquiring the major revision opportunity from

Journal of Operational Research Society (JORS) in December, 2015. This work is

linked to the interface between operation management and social responsibility which is a hot topic recently. Introducing social responsibility in product development and promotional strategies can affect the cost and demand of a firm’s product, as well as the social impact. We explore whether a for-profit firm should employ a purchased-triggered donations (PRD) strategy by proposing an analytical model in which a profit-maximizing firm decides on the product price and its donation amount to the society. Two dimensions of consumer behavior are considered – price and

(6)

social concern. We identify two common forms of socially responsible action: profit-maximizing ethics, through which socially responsible behavior can actually improve firm profitability, and costly philanthropy, whereby firms engage in socially responsible behavior for their ethical and social value, even if their profitability is consequently reduced. We develop managerial insights by characterizing and exploring the optimal and equilibrium solutions. Contrary to conventional wisdom, our analysis shows that both monopolistic and duopolistic firms prefer not to engage in socially responsible behavior through PRD. On the other hand, if an upper limit is applied on the PRD amount, then competing firms may find themselves in equilibrium with either profit-maximizing ethics or costly philanthropy. In this sense, we conclude that PRD-based corporate social responsibility (CSR) behavior is best described as a competitive necessity for firms engaged in competition.

Work-in-Process

This project not just support the publication of two papers aforementioned, but also provide aid for the paper “To grow or survive: Managing opportunity-driven supply chains by startups”. We submit this work to POM in 2014 and unfortunately we get rejection due to insufficient effort to address the importance of this research. With the reviewer’s constructive comments and the funding support from this project, we totally revise the manuscript now entitled “How much demand should be induced: partial sourcing to entrepreneurs subject to capacity shortage”. In the following, we will first show our revised work and eventually point out the difference between the original manuscript and the revised one in the last paragraph.

1. Introduction

Entrepreneurship is one of the key drivers in the development of new products to explore new markets and exploit potential demand. In recent years, firms such as original equipment manufacturers (OEMs) rely less on internal resources, but instead, more of them have increasingly relied on external start-ups for the development of new products that are required for the exploitation of new business opportunities (Billington and Davidson 2013, Hora and Dutta 2013, Krishnan 2013, Xiao and Gaimon 2013). It is often the case in business-to-business markets that an entrepreneurial technology provider (entrepreneur, hereafter) introduces and sells innovative components to buying firms (buyers, hereafter) such as OEMs that also outsource to experienced supply chain counterparts, i.e., partial sourcing, to stabilize market supply of new products rather than outsourcing only to the start-up, i.e., complete sourcing (Anderson and Parker 2013, Erat and Kavadias 2006, Gray et al. 2009, Ü lkü and Schmidt 2011).

(7)

Despite the popularity and benefits of using partial sourcing (or multiple suppliers, e.g., matching supply with demand, cost reduction, quality improvement; see Cachon and Terwiesch 2013, Chopra and Meindl 2012), entrepreneurs who provide component technologies, on the other hand, usually concern that partial sourcing induces upstream competition in obtaining buyers’ orders and then harm their performance (Arend and Wisner 2005, Wang and Shin 2015). However, many entrepreneurs in high-tech sectors (most notably consumer electronics) may share their technological knowhow with the established supply chain counterparts to help guarantee their buyers’ need for a quick new product launch that typically require reliable supply of components (Erat 2006, Savva and Scholtes 2014). This is because with partial sourcing, an entrepreneur can leverage its growth utilizing reliable suppliers’ cost advantage and scaling ability when that entrepreneur exploits the buyers’ demand yet may be unable to fulfill their orders.

We offer the following example of Apple’s introduction of the iPhone: In 1997, start-up TPK Holding’s founder, Michael Chiang, sensed that multipoint touchscreens would rapidly expand the boundaries of consumer electronics. Once TPK’s touchscreen technology was functional, TPK focused on developing touch panels for mobile phones and in 2003 introduced its technology to the leading mobile phone producers. However, only Apple, which had been in the early stages of developing its first iPhone, was interested. TPK became Apple’s first supplier of multipoint touchscreens, which soon became the main feature across Apple’s product line of iPads and iPods. However, TPK alone could not supply Apple due to its limited capacity as well as low production yield rate as a start-up. Apple then asked reliable and established suppliers such as Wintek to supply the same touchscreen to keep pace with the growing demand of iPhone and iPad (Walker 2007a, 2007b, 2007c). Ultimately, TPK’s technological knowhow enhanced the demand for Apple’s end products and consequently TPK benefited from Apple’s partial sourcing strategy to survive during the early stage of market entry, leading to subsequent growth (Flannery 2012, Ganeshan 2010, Slivka 2010). This example is one among many, e.g., Erat (2006) on aircraft manufacturing, Ö zer and Raz (2011) on hearing implants, Savva and Scholtes (2014) on biotechnology and pharmaceuticals, and Paik and Zhu (2014) on smartphones. These real world examples describe business context that is the focus of our paper: an entrepreneur sells innovative components to buyers and a partial-sourcing supply chain can serve as a mechanism for growing their business under the threat of capacity shortages.

The rationale for the attractiveness of partial sourcing to entrepreneurs is as follows. Buyers who face end-consumers launch some marketing campaigns to create buzz and drum up interest for innovative product features derived from an

(8)

entrepreneur’ advanced component technology. The interest is further enhanced by announcing the increase of order quantities of new components to the market to expand potential demand. We refer this phenomenon as the order-induced effect. This effect increases the likelihood of supply from the entrepreneur being insufficient to meet the inflated demand from the buyers. Furthermore, the order-induced effect may be even generated prior to the production of components so it is not necessarily restricted by the entrepreneur’s available capacity (Sapra et al. 2010). Partial sourcing can ease the negative effect of capacity shortages on exploiting the market size of new products; the end outcome is more demand.

We consider a two-stage supply chain that an entrepreneur who is an upstream supplier may not have sufficient capacity to meet demand from buyers who are downstream players. The strategic sourcing in this research refers to an entrepreneur’s decision on whether adopting the complete sourcing, e.g., buyers’ demand only from this entrepreneur, or adopting the partial sourcing, e.g., buyers’ demand from this entrepreneur and also from other suppliers who have the ability to compensate the entrepreneur’s capacity shortage.

Consider a stylized supply chain with an entrepreneur (s), who plays the role of an upstream supplier, and n identical downstream buyers (b), each of them facing an uncertain demand D = d + β(q) where q is each buyer’s order quantity from the entrepreneur. Following the conventional practice in operations literature (see, e.g., Porteus 2002), d is a random variable with mean μ, a probability density function f(∙), and a cumulative density function F(∙) that is strictly increasing, differentiable, and F(0) = 0; and, all firms are risk-neutral and profit maximizers. If β(∙) is an increasing function, d can be viewed as the underlying demand for each buyer. We use the common convention in the supply chain literature that the supplier is female and the buyer is male, i.e., in what follows, we refer to the entrepreneur as ‘she’ and the buyer as ‘he’. All proofs appear in the Appendix.

In practice, the entrepreneur may not only plays a passive role of supplying components to the buyers, captured by the stochastic component d of end-product demand, but also plays an active role of being capable of increasing each buyer’s end-product demand through its superior technological knowhow, captured by the deterministic component β(∙) of end-product demand. Specifically, function β(q) depicts how each buyer’s demand for the end-product can grow up by increasing the order quantity q from the entrepreneur (i.e., the order-induced effect). In other words, the entrepreneur’s output q would signal how much she can grow the buyer’s end-product market. Let β(∙) be a non-decreasing concave function to depict diminishing return from the entrepreneur’s output, as do Alcacer and Oxley (2014) and Gaimon and Bailey (2013), so that 0 ≤ β′(∙) < 1 and its maximum β̅ can be

(9)

reached for any q ≥ q̅ ≡ (β′₎−1_{(0) and β̅ < q̅. Throughout this paper, we use ′ to}

denote a first-order derivative and ′′ to denote a second-order derivative when there is no possibility of ambiguity. A suitable choice of the β(∙) function can capture a variety of scenarios. For instance, if increasing demand beyond the entrepreneur’s capacity cannot generate additional demand, β(q) may be chosen so that it is constant for large values of q. The curvature of β(∙) may be interpreted as the intensity of the order-driven effect of demand growth.

Suppose that the entrepreneur’s capacity may be insufficient to meet the order quantity from the buyers so that the entrepreneur will ration the available capacity according to the fraction of order quantity accounted for by each buyer, following the conventional rationing game setting in supply chain literature (see, e.g., Lee et al. 1997). For instance, if a buyer accounts for 15% of the total order quantity from all the buyers, it will receive 15% of the available capacity.

We assume that the entrepreneur’ available capacity to supply the need of buyers can take on two quantities due to limited experience and newness to the market: a constant K (i.e., a capacity limit) with probability α ∈ [0,1) and ∞ with probability 1 − α. That is, with probability α there will be a capacity shortage and with probability 1 − α there will be adequate supply without any capacity limit. In other words, buyer i, i = 1, ⋯ , n, orders q_i from the entrepreneur and then receives q_i with probability 1 − α and Kq_i/ ∑n_i=1q_i with probability α. To mainly focus on the entrepreneur’s point of view, in this paper, the buyers are assumed to be identical, as do many model-based studies including Lee at al. (1997), so that they will make exactly ordering decisions: q = q₁ = q₂ = ⋯ = q_n. Then, each buyer’s allocation is η = K/n while capacity shortages occur. Let β = β(η) such that η > β. Note that η is assumed to be less than the order quantity placed by each buyer so that capacity shortages are meaningful in our framework.

As a buyer attempts to convince potential consumers to buy his new product that the key feature is the adoption of the entrepreneur’s components, the new product’s demand potential may be limited by the entrepreneur’s available capacity. Specifically, if the entrepreneur’s orders beyond her available capacity cannot generate additional demand, demand growth can be modeled as max{β(q), β(η)}, called limited order-induced effect. That is, the new product’s potential market size is decided by the entrepreneur’s available capacity, i.e., the fraction of the order quantity buyers can receive. If the new product’s potential market size is determined by the buyer’s order quantity that may be less than her available capacity, demand growth can be straightly modeled as β(q), called unlimited order-driven effect.

There are two conceivable supply chain configurations in the supply chain and product development literatures: complete sourcing (CS) and partial sourcing (PS).

(10)

The former configuration is a straightforward case that the entrepreneur is the only supplier of critical components for each buyer. The latter one is that each buyer outsources only a fraction of needed order quantity from the entrepreneur. We use r to represent a buyer’s order quantity from the other supplier that posses no technological knowhow to generate the order-induced effect but has enough capacity to supply. We assume that this type of established suppliers is not playing any strategic role in our framework.

Entrepreneurs and buyers can engage in joint development of a new product in numerous ways. To begin making a contribution to the emerging literature on operations-entrepreneurship interface, we consider that the entrepreneur and the buyers are currently in a revenue sharing agreement wherein each buyer gets a fraction ϕ ∈ (0,1] of the total revenue, whereas the entrepreneur receives 1 − ϕ of the revenue, in line with the supply chain literature (Cachon and Lariviere 2005). Revenue sharing has been shown to be a useful mechanism for coordination in the product development literature that deals with the co-development of new products between different supply chain parties (Bhaskaran and Krishnan 2009). For example, Apple uses revenue-sharing contracts and shares 70% of the sales revenue with its app developers, whose products significantly increase the value of Apple’s products such as iPad and iPhone (Wang and Shin 2015).

Under revenue sharing, the entrepreneur is encouraged to allow partial sourcing in the supply chain configuration; specifically, in this agreement, the fraction 1 − ϕ of the buyer’s revenue generated from outsourcing to the established supplier that he will share with the entrepreneur. As ϕ decreases, the entrepreneur obtains more revenue share, which then provides her with a stronger incentive to favor partial sourcing for further growth; in contrast, the entrepreneur favors complete sourcing.

There are two stages in the model. In the first stage, an entrepreneur identifies a business opportunity to be exploited due to the order-induced effect, leading to: the entrepreneur and the potential buyers employ an agreement to form a supply chain of launching new products to end customers: 1) each buyer shares 1 − ϕ portion of his revenue to the entrepreneur, and 2) which sourcing structure, complete or partial, is implemented. In the second stage, each buyer must select an order quantity, determining its stock level of the entrepreneur’s components. The uncertain demand component d is then realized and revealed to the buyers, along with the entrepreneur’s available capacity. If demand exceeds the stock level, the buyer faces a backorder (or backlogging) penalty cost for having too little inventory. This penalty cost includes the lost-opportunity value of delayed revenue, and/or the cost to place an expedited order, and/or the loss of goodwill and future sales. If the stock level exceeds demand, the buyer must cover the holding cost for the remaining inventory.

(11)

We use λ to represent the backorder penalty cost per unit and h to represent the cost per unit of excess inventory (overage cost). The product price (i.e., revenue per unit sold by buyers) is p per unit. The entrepreneur’s production cost is c_s per unit and each buyer’s production cost is c_b per unit. Each buyer pays the entrepreneur a given cost w per unit ordered and other suppliers v per unit ordered, respectively. Without loss of generality, we assume that one product unit requires exactly one unit of the supplier’s critical component (since we can scale the number of components required for one product unit by adjusting the total cost). In line with the analytical literature (e.g., Anand and Girotra 2007, Druehl and Schmidt 2008, Goyal and Netessine 2007), in our context, the entrepreneur identifies the business opportunity in the beginning. Consequently, the market-entry cost is sunk and disregarded.

2. Problem Formulation

We are now ready to formulate each buyer’s and entrepreneur’s expected profit functions. To that end, let L₁(q, r) be the expected backorder at the end of period if there is no capacity shortage:

L₁(q, r) = ∫ (x + β(q) − q − r)f(x)dx

∞ q+r−β(q)

.

When there is a capacity shortage, the expected backorder at the end of period changes under different order-induced effects. If the potential market size is restricted from the available supply from the entrepreneur, i.e., limited order-induced effect, it is L₁(η, r); otherwise, i.e., unlimited order-induced effect, the expected backorder at the end of period is:

L₂(q, r) = ∫ (x + β(q) − η − r)f(x)dx

∞ η+r−β(q)

.

Let I₁(q, r) be the expected inventory on-hand at the end of period if there is no capacity shortage:

I1(q, r) = ∫ (q + r − x − β(q))f(x)dx. q+r−β(q)

0

When there is a capacity shortage, the expected inventory on-hand at the end of period changes under different order-induced effects. If the order-induced effect is limited by the available capacity of the entrepreneur, it is I₁(η, r) . If the order-induced effect is unlimited, the expected inventory on-hand at the end of period is:

I2(q, r) = ∫ (η + r − x − β(q))f(x)dx. η+r−β(q)

0

(12)

order-induced effect,

π_b,1(q, r, α) = (1 − α)ϕpE[d + β(q)] + αϕpE[d + β(η)]

− (1 − α)[λL₁(q, r) + hI₁(q, r) + (c_b+ w)q + (c_b+ v)r] − α[λL₁(η, r) + hI₁(η, r) + (c_b+ w)η + (c_b+ v)r]; under the unlimited order-induced effect,

π_b,2(q, r, α) = ϕpE[d + β(q)]

− (1 − α)[λL₁(q, r) + hI₁(q, r) + (c_b+ w)q + (c_b+ v)r] − α[λL₂(q, r) + hI₂(q, r) + (c_b+ w)η + (c_b+ v)r].

In both profit functions, the former of the last two terms represents the expected cost given no shortage occurring while the latter term for the capacity shortage happening. Other terms represent the revenue.

The entrepreneur’s expected profit function is thus formulated as: under the limited order-induced effect,

π_s,1(q, α) = n[(1 − α)[(1 − ϕ)pE[d + β(q)] + (w − c_s)q] + α[(1 − ϕ)pE[d + β(η)] + (w − c_s)η]]; under the unlimited order-induced effect,

π_s,2(q, α) = n[(1 − α)[(1 − ϕ)pE[d + β(q)] + (w − c_s)q] + α[(1 − ϕ)pE[d + β(q)] + (w − c_s)η]].

The supply chain’s total expected profit function is therefore: under the limited order-induced effect,

Π₁(q, r) = nπb,1(q, r, α) + πs,1(q, α);

under the unlimited order-induced effect,

Π2(q, r) = nπb,2(q, r, α) + πs,2(q, α).

The supply chain is considered coordinated if the ordering decisions, q and r, in a decentralized supply chain are identical to those in a centralized supply chain where there is a central planner to maximize the total supply chain profit, i.e., the optimal order quantities coincide.

3. Analysis of Order-Induced Effect

To reflect the buyer’s reality that failing to match demand with supply is costly, we need a mild assumption in the subsequent sections:

Assumption 1. (a) ϕp > λ; (b) λ > v + c_b; (c) w > v.

Assumption 1a rules out some extreme cases where the entrepreneur’s component supply is not worthwhile in the buyer’s perspective. Assumptions 1bc hold whenever that partial sourcing can be an alternative option while the established suppliers who cannot create the order-driven effect have a cost advantage over the entrepreneur (see e.g., Ö zer and Raz 2011). This assumption rule out the trivialities that it is never profitable in expectation to exploit the business opportunity through the

(13)

entrepreneur’s supply and partial sourcing to different types of suppliers, the main focus of our paper, and it is thus satisfied in the great majority of real world situations. (It is notable that many analytical studies in the literature implicitly make similar assumptions by scaling demand functions and operational costs).

We now analyze this two-stage model. In the first stage, given the value of ϕ the supply chain decides whether sourcing structure, complete or partial, to employ. In the second stage, each buyer makes the order quantity decision prior to ascertaining the realized demand under the threat of capacity shortage in the entrepreneur’s side. We obtain the overall solution through backward induction, where we solve our model from the buyer’s perspective first.

3.1 Buyers’ Problem

Given the type of order-induced effects, j ∈ {1, 2}, where j = 1 refers the limited order-induced effect and j = 2 refers to the unlimited order-induced effect, in the second-stage of our model that takes place if complete sourcing (CS) is employed, each buyer selects the stock level q that maximizes his expected profit max_q>0,r=0 π_b,j(q, r, α) . We characterize each buyer’s optimal-order-quantity decision under complete sourcing in Lemma 1.

Lemma 1. If complete sourcing is employed, then

(a) under the limited order-induced effect, each buyer’s optimal order quantity, q₁∗(CS), satisfies the following equality:

F(q − β(q)) =(ϕp − λ)β

′_{(q) + λ − c} b− w

(λ + h)[1 − β′_(q)] ;

(b) under the unlimited order-driven effect, each buyer’s optimal order quantity, q∗₂(CS), satisfies the following equality:

(1 − α)(1 − β′_{(q))F(q − β(q)) − αβ}′_{(q)F(η − β(q))}

=(ϕp − λ)β

′_{(q) − (1 − α)(c}

b+ w − λ)

λ + h .

If, on the other hand, partial sourcing (PS) is employed, then the buyer selects stock levels q and r from the entrepreneur and established supplier, respectively, aimed to maximize his expected profit expressed by max_q,r>0 π_b,j(q, r, α) , the objective function of which is concave. We characterize the buyer’s optimal-order-quantity decision under partial sourcing in Lemma 2.

Lemma 2. If partial sourcing (PS) is employed, then

(a) under the limited order-induced effect, each buyer’s optimal order quantity pair, q₁∗(PS) and r₁∗_{(PS), satisfies the following system of equations:}

(14)

F(q + r − β(q)) =(ϕp − λ)β ′_{(q) + λ − c} b− w (λ + h)[1 − β′_(q)] and (1 − α)F(q + r − β(q)) + αF(η + r − β(η)) = λ − cb− v λ + h ;

(b) under the unlimited order-induced effect, each buyer’s optimal order quantity pair, q∗₂(PS) and r₂∗_{(PS), satisfies the following system of equations:}

(1 − α)(1 − β′_{(q))F(q + r − β(q)) − αβ}′_{(q)F(η + r − β(q))} =(ϕp − λ)β ′_{(q) − (1 − α)(c} b+ w − λ) λ + h and (1 − α)F(q + r − β(q)) + αF(η + r − β(q)) =λ − cb− v λ + h .

It is straightforward to see that each buyer favors partial sourcing over complete sourcing because this supply chain configuration provides more flexibility in a general sense. Theoretically speaking, the optimal solution for the buyer’s problem under complete sourcing, max_q>0,r=0 π_b,j(q, r, α), is a suboptimal solution for the buyer’s problem under partial sourcing, max_q,r>0 π_b,j(q, r, α) so the following result always holds:

Remark 1. Each buyer should favor partial sourcing over complete sourcing.

We also can argue, roughly, that the buyer can benefit from partial sourcing since the other supplier’s procurement cost is lower than that of the entrepreneur; conversely, the buyer does not want to give up the entrepreneur because it can help the buyer to increase demand due to the order-induced effect.

Now move to the entrepreneur’s view of point regarding the order quantity placed by each buyer. By ordering more from the entrepreneur, the buyer can expand the market size while the size might be restricted by the entrepreneur’s available capacity. Yet we show, in Lemma 3a, that the entrepreneur receives a smaller order under partial sourcing due to the market potential constrained by its capacity limit. However, Lemma 3b suggests that the unlimited order-induced effect is possible to lead to the outcome that the entrepreneur receives a larger order under partial sourcing, even though it faces a procurement cost disadvantage against the established suppliers.

Lemma 3. Let δ ≡ G(q∗₂(CS), r₂∗_{(PS)) − G(q} 2

∗_{(CS), 0) where G(q, r) = −(1 −}

α)(1 − β′_{(q))F(q + r − β(q)) + αβ}′_{(q)F(η + r − β(q)). Then}

(15)

the entrepreneur is smaller under partial sourcing than under complete sourcing, or both equal, q₁∗(PS) ≤ q∗₁_(CS);

(b) under the unlimited order-induced effect, if δ ≤ 0, each buyer’s optimal order quantity from the entrepreneur is smaller under partial sourcing than under complete sourcing, or both are equal, q₂∗(PS) ≤ q₂∗(CS); equivalently, δ > 0 leads to q∗₂(PS) > q∗₂_(CS).

The condition in Lemma 3b is general. In order to distill insights, we use a uniform distribution to describe random variable d in the demand function. So we can observe a pattern, in Corollary 1, that the likelihood of δ > 0 is increasing in α; roughly speaking, larger order under partial sourcing may take place when the order-induced effect is capacity-unconstrained and the capacity shortage risk is sufficiently high.

Corollary 1. Suppose that d is uniformly distributed with support [0, c]. Under the unlimited order-induced effect, then

(a) each buyer’s optimal order quantity from the entrepreneur is smaller under partial sourcing than under complete sourcing, or both are equal, q∗₂(PS) ≤ q∗₂(CS) if 1 − α ≥ β′(q₂∗_(CS));

(b) otherwise, each buyer’s optimal order quantity from the entrepreneur is greater under partial sourcing than under complete sourcing, q∗₂(PS) > q∗₂_(CS).

3.2 Entrepreneur’s Problem

In our framework, prior to reaching the agreement of sourcing structure and revenue sharing, the entrepreneur may anticipate the preference from each buyer as she maximizes her expected profit from the buyers based on which sourcing structure is used. Remark 1 demonstrates that the buyer always favors partial sourcing as the supply chain configuration of launching a new product; however, Lemma 3 suggests that this monotonic result does not hold in the entrepreneur’s point of view. Our next theorem formally characterizes the entrepreneur’s preference for sourcing structure. Theorem 2. (a) Under the limited order-induced effect, the entrepreneur should favor complete sourcing. (b) Under the unlimited order-driven effect, if δ ≤ 0 , the entrepreneur should favor complete sourcing; otherwise, the entrepreneur should favor partial sourcing.

Theorem 2a formalizes the intuitive property that the entrepreneur should prefer complete sourcing if demand is limited by the entrepreneur’s available capacity. This result is intuitive because the potential market size is not able to expand due to

(16)

her capacity limit while shortages take place. Theorem 2b shows that if capacity shortages do not impact the potential market size (i.e., unlimited order-induced effect), unlike the limited order-induced effect, partial sourcing to the entrepreneur may be preferable under δ > 0. Corollary 1 explicitly suggests, under a uniform demand uncertainty, that the outcome is likely to take place for a sufficiently high risk of capacity shortages. We therefore obtain:

Remark 2. The unlimited order-induced effect is possible to lead to the outcome that the entrepreneur favors partial sourcing, which is impossible to occur under the limited order-induced effect.

To summarize our analysis in this section: The main conclusion is that complete sourcing is not always the preference to the entrepreneur (Remark 2). By specifying a uniform demand uncertainty, we have the result (Corollary 1) that a high rick of capacity shortages makes the entrepreneur prefer partial sourcing when that risk does not limit the potential market size. The interesting observation here is that the entrepreneur and buyers favor partial sourcing since a high shortage risk makes each buyer be motivated to inflate his order quantity from the entrepreneur to expand the potential market size of new products; this behavior does not occur when the potential market size depends on the entrepreneur’s available supply to the buyers rather than the orders from them.

4. Capacity Shortage Effect

In this section we examine another research question: How the threat of capacity shortages impacts the entrepreneur’s order quantities placed by the buyers? We start by considering that the entrepreneur’s available capacity is infinite, i.e., α = 0. The optimal ordering decision of each buyer can be characterized by specifying α = 0 in Lemmas 1 and 2:

Corollary 2. Suppose that α = 0, i.e., there is no capacity shortage risk facing the entrepreneur. If complete sourcing is employed, then each buyer’s optimal order quantity, q̂(CS), satisfies the following equality:

β′(q) = w − v ϕp − v − cb

(might not ture, use below) F(q − β(q)) =(ϕp − λ)β

′_{(q) + λ − c} b− w

(λ + h)(1 − β′_(q))

(17)

Corollary 3. Suppose that α = 0, i.e., there is no capacity shortage risk facing the entrepreneur. If partial sourcing is employed, then each buyer’s optimal order quantity pair, q̂(PS) and r̂(PS), satisfies the following system of equations:

β′(q) = w − v ϕp − v − cb

and F(q + r − β(q)) =λ − v − cb λ + h . no matter whether the order-induced effect is limited or not.

We further find that the entrepreneur receives a smaller order under partial sourcing in the absence of capacity shortages.

Lemma 4. In the absence of capacity shortages i.e., α = 0, then each buyer’s optimal order quantity from the entrepreneur is smaller under partial sourcing than under complete sourcing, or both equal, q̂(PS) ≤ q̂(CS).

Because the entrepreneur’s profit function in terms of different types of the order-induced effect is increasing in the order quantity received, we can thus conclude:

Theorem 3. In the absence of capacity shortages, i.e., α = 0, the entrepreneur should favor complete sourcing over partial sourcing.

With Corollary 2 comparing to Lemma 1, we can examine, under complete sourcing, whether the entrepreneur’s order quantities placed by the buyers will be inflated in the presence of capacity shortages. This can be seen in the following lemma:

Lemma 5. If complete sourcing is employed, then

(a) under the limited order-induced effect, q₁∗(CS) = q̂(CS); (b) under the unlimited order-induced effect, q₂∗(CS) > q̂(CS).

By Lemma 5, it can be seen, under complete sourcing, that the order inflation due to capacity shortage only takes place when the potential market size is upon the entrepreneur’s received orders rather than her available supply.

Theorem 4. Suppose that complete sourcing is employed. If the order-induced effect is capacity-constrained, capacity shortage risk does not impact the entrepreneur‘s order quantities from the buyers. If the order-induced effect is capacity-unconstrained, the entrepreneur has more order quantities from the buyers in the presence of capacity

(18)

shortages than that in the absence of capacity shortages.

Now consider the partial sourcing case. With Corollary 3 comparing to Lemma 2, we can examine, under partial sourcing, whether the entrepreneur’s order quantities placed by the buyers will be inflated in the presence of capacity shortages. This can be seen in the following lemma:

Lemma 6. Let

θ ≡ (c_b+ w − λ) + [(λ + h) (1 − β′_{( q̂(PS))) F ( q̂(PS) + r}

2∗(PS) − β( q̂(PS)))].

If partial sourcing is employed, then

(a) under the limited order-induced effect, either {q∗₁(PS) ≤ q̂(PS), r₁∗_{(PS) ≥ r̂(PS)}}

or {q₁∗(PS) > q̂(PS), r₁∗_{(PS) < r̂(PS)} occurs;}

(b) under the unlimited order-induced effect, q∗₂(PS) ≥ q̂(PS) if (ϕp − λ)β′(q̂(PS)) − θ ≥ 0; otherwise, q∗₂(PS) < q̂(PS).

Lemma 6 shows that under partial sourcing, the buyers may deflate the order quantities from the entrepreneur, which is impossible to occur under complete sourcing. In other words, in the presence of capacity shortages, order quantities from the entrepreneurs may be deflated.

Remark 3. Partial sourcing is possible to lead to the outcome that the entrepreneur has less order quantities from the buyers in the presence of capacity shortages than that in the absence of capacity shortages, which is impossible to occur under complete sourcing.

In sum, we have shown that, with partial sourcing, the presence of capacity shortages may lead to the order deflation to the entrepreneur, which is impossible to occur with complete sourcing. In addition, we find that capacity shortage cannot either inflate or deflate the entrepreneur’s order quantities placed by the buyers while complete sourcing is implemented.

5. Summary of revised manuscript

As for the work submitting to POM, we follow reviewers’ suggestion to revise our manuscript with the following key directions.

(1) We need to consider the transfer fee when we discuss touch the technology management area in which a firm who owns technology will get financial reward from firms who use its technology for production.

(19)

(2) We need to address a practical issue that an entrepreneur will have the capacity shortage risk.

(3) We need to have the practical examples to support our research.

Reference

Alcacer, J., J. Oxley. 2014. Learning by supplying. Strategic Management Journal

35(2) 204–223.

Anand, K. S., K. Girotra. 2007. The strategic perils of delayed differentiation.

Management Science 53(5) 697–712.

Anderson, E. G., G. G. Parker. 2013. Integration of global knowledge networks.

Production and Operations Management 22(6) 1446–1463.

Anupindi, R., R. Akella. 1993. Diversification under supply uncertainty. Management

Science 39(8) 944–963.

Arend, R. J., J. D. Wisner. 2005. Small business and supply chain management: Is there a fit? Journal of Business Venturing 20(3) 403–436.

Arya, A., B. Mittendorf, D. E. M. Sappington. 2008. The make-or-buy decision in the presence of a rival: Strategic outsourcing to a common supplier. Management

Science 54(10) 1747–1758.

Bhargava, H. K., B. C. Kim, D. Sun. 2013. Commercialization of platform technologies: Launch timing and versioning strategy. Production and

Operations Management 22(6) 1374–1388.

Bhaskaran, S., V. Krishnan. 2009. Effort, revenue, and cost-sharing in contractual new product development. Management Science 55(7) 1152–1169.

Billington, C., R. Davidson. 2013. Leveraging open innovation using intermediary networks. Production and Operations Management 22(6) 1464–1477.

Cachon, G., M. Lariviere. 2005. Supply chain coordination with revenue sharing: Strengths and limitations. Management Science 51(1) 30–44

Cachon, G., C. Terwiesch. 2013. Matching Supply with Demand: An Introduction to

Operations Management, 3rd edition. New York, NY: McGraw-Hill.

Chopra, S., P. Meindl. 2012. Supply Chain Management: Strategy, Planning, and

Operation, 5th edition. Upper Saddle River, NJ: Pearson.

Druehl, C. T., G. M. Schmidt. 2008. A strategy for opening a new market and encroaching on the lower end of the existing market. Production and

Elmaghraby, W. J. 2000. Supply contract competition and sourcing policies.

Manufacturing and Service Operations Management 2(4) 350–371.

(20)

Doctoral Dissertation. College of Management, Georgia Institute of Technology, Atlanta, GA.

Erat, S., S. Kavadias. 2006. Introduction of new technologies to competing industrial customers. Management Science 52(11) 1675–1688.

Flannery, R. 2012. Bobbing with Apple: Taiwan touchscreen champion TPK faces test ahead. Forbes, May 23.

Ganeshan, R. 2010. The iPhone 4 supply chain. Operations Buzz, November 28. Gaimon, C., J. Bailey. 2013. Knowledge management for the entrepreneurial venture.

Goyal, M., S. Netessine. 2007. Strategic technology choice and capacity investment under demand uncertainty. Management Science 53(2) 192–207.

Gray, J. V., B. Tomlin, A.V. Roth 2009. Outsourcing to a powerful contract manufacturer: The Effect of learning-by-doing. Production and Operations

Management 18(5) 487–505.

Hora, M., D. K. Dutta. 2013. Entrepreneurial firms and downstream alliance partnerships: Impact of portfolio depth and scope on technology innovation and commercialization success. Production and Operations Management 22(6) 1389–1400.

Joglekar, N., M. Lévesque. 2013. The role of operations management across the entrepreneurial value chain. Production and Operations Management 22(6) 1321–1335.

Kapoor, R., R. Adner. 2012. What firms make vs. what they know: How firms' production and knowledge boundaries affect competitive advantage in the face of technological change. Organization Science 23(5) 1227–1248.

Krishnan, V. 2013. Operations management opportunities in technology commercialization and entrepreneurship. Production and Operations

Management 22(6) 1439–1445.

Ö zer, Ö ., G. Raz. 2011. Supply chain sourcing under asymmetric information.

Paik, Y., F. Zhu. 2014. The impact of patent wars on firm strategy: Evidence from the globalsmartphone market. Harvard Business School Working Paper.

Phan, P., C. Chambers. 2013. Advancing theory in entrepreneurship from the lens of operations management. Production and Operations Management 22(6) 1423– 1428.

Porteus, E. 2002. Foundations of Stochastic Inventory Theory. Stanford, CA: Stanford University Press.

Sapra, A., V. A. Truong, R. Q. Zhang. 2010. How much demand should be fulfilled?

(21)

Savva, N., S. Scholtes. 2014. Opt-out options in new product co-development partnerships. Production and Operations Management 23(8) 1370–1386. Shepherd, D. A., H. Patzelt. 2013. Operational entrepreneurship: How operations

management research can advance entrepreneurship. Production and

Slivka, E. 2010. Apple's iPad touch panel supply woes continue, shifts production from TPK to Wintek. MacRumors, March 26.

Snyder, L. V., Shen, Z. M. 2011. Fundamentals of Supply Chain Theory. Hoboken, NJ: John Wiley & Sons.

Ü lkü, S., Schmidt, G. 2011. Matching product architecture with supply chain configuration. Production and Operations Management 20(1) 16–31.

Ulrich, K. T., D. J. Ellison. 2005. Beyond make-buy: Internalization and integration of design and production. Production and Operations Management 14(3) 315– 330.

Walker, G. 2007a. Touch and the Apple iPhone. Veritas et Visus Touch Panel, February.

Walker, G. 2007b. The impact of the Apple iPhone on the touch-panel industry. SID

Information Display, July.

Walker, G. 2007c. Last word: Who the heck is Balda? Veritas et Visus Touch Panel, July.

Wang, J., H. Shin. 2015. The impact of contracts and competition on upstream innovation in a supply chain. Production and Operations Management 24(1) 134–146.

Wu, X., A.-M. Knott. 2006. Entrepreneurial risk and market entry. Management

Science 52(9) 1315–1330.

Xiao, W., C. Gaimon. 2013. The effect of learning and integration investment on manufacturing outsourcing decisions: A game theoretic approach. Production

(22)

1

科技部補助專題研究計畫出席國際學術會議心得報告

日期： 105 年 1 月 29 日

計畫編

號

MOST 103－ 2410 － H － 004 －

107 －

計畫名

稱

探索供應鏈架構下企業機會之分析:從採購與生產的角度

出國人

員姓名

陳立民

服務機

構及職

稱

政治大學企管系助理教授

會議時

間

2015 年 10 月

21 日至

2015 年 10

月 23 日

會議地

點

西班牙賽維亞(Seville)

會議名

稱

2015 IESM 研討會

發表題

目

How Does a Responsible Supplier Control its Production under

the Opportunity of Demand Expansion?

(23)

2 中文本文摘要越來越多供應鏈上的參與者將道德意識結合於企業營運上很重要的一部份。本研究藉由供應商的角度來探討其藉由一個創新性的行為，轉型為道德供應商，並生產負責任的產品(零組件，原物料等等……)來提供給供應鏈的下游買家。此轉型最大的好處是可探索出一群全新的商業機會。換句話說，其不只可吸引原本以價格為優先的下游買家，同時也可吸引到有道德意識的一群全新的買家。藉由計量模型，我們分析此一供應商的生產數量之結構。同時本研究也會將此分析的結構藉由市場上意識到的此全新的商業機會大小來做更詳細的敘述。英文本文摘要

Increasing supply chain players seek to run business in a responsible way. This work stands from a supplier’s point of view by considering a two-stage supply chain structure in which a responsible supplier produces a responsible component (or raw material, ingredient) an entrepreneurial process and exploits the business opportunity (e.g., market expansion) to sell it to downstream buyers. Buyers’ market is classified by two: one for price-sensitive and one for responsible-conscious. We develop quantitative models to analyze such a supplier’s production structure, interplaying with the attractiveness to the business opportunity exploitation.

(24)

3

本文:

1. 會議敘述與目的

本屆會議 ( 2015 International Conference on Industrial Engineering and Systems Management) 於 2015 年 10 月 21 號至 23 號於西班牙賽維亞進行一年一度的年會。今年的會議定調 “The Rod Ahead: Understanding Challenges and Grasping Opportunities in Industrial and Systems Engineering”。針對議題部分，細分了多項專業的討論 session 來做進行，其中包括: Innovation management and entrepreneurship、Production planning and scheduling 和 Warehouse and inventory management。

2. 發表之過程

本人是被安排於 10 月 22 號上午 8:30-10:00am 的場次做文章發表，同時也擔任此 session 的主席。有鑑於此次會議的主軸涵蓋Entrepreneurship，Supply Chain Management，與 Ethics。發表的文章剛好是此三個領域的一個結合與運用。發表首先利用 2013 年於 Bangladesh 發生的成衣工廠倒塌的事件與之前富士康 (foxconn)員工跳樓的案件來點起本研究目的。在成衣工廠的事件當中，大樓的倒塌造成的傷亡，固然引起社會的關注，但更重要的是媒體跟輿論開始注意到原來在 Bangladesh 絕大部分在此類似的成衣場裡工作的女工們，皆有薪資被剝削的嚴重問題。因此在一些國家裡(例如英國)便發生了 NGO 團體組織與消費者集體發動拒買一些國際知名品牌(如 NIKE， Mark Spencer， Primark)的衣服，探究根本原因是這些大廠皆是委託在 Bangladesh 這些成衣加工廠製做其所需的衣飾。再進入到富士康的例子當中，前陣子的跳樓事件，引起了人權組織與輿論抨擊富士康與蘋果公司。從這兩個案例當中，我們看到了當上游供應商出了問題之後，終端消費者與媒體不只將矛頭指向他們，同時這些知名具有市場影響力的下游品牌商與零售商，所受到的指責更是強烈。因此本研究鑑於越來越多的末端消費者排斥購買沒有道德責任的商品，進而導致這些品牌商與零售商開始尋求從道德的供應商，期從他們那邊，獲取藉由道德化的生產過程產出的原物料或零組件等等。藉由供應商的一個所謂創新的創業過程 (entrepreneurial process)，來提供負有道德責任的原物料與零組件給下游買家。但是此時供應商的轉型會面臨一些難處，如以下列舉。 (1)雖然可探索新的商業機會賣給有道德意識的買家，但不知帶來的市場效益大小為何，進而其會對供應商的生產數量計畫有何影響。其中市場效益尚無法預期是因為這樣的研究目前皆缺乏一些具體的資料來做評估。

(26)

5

(2)供應商若藉由道德的生產行為，一般預計會比原來的生產行為產生較高的生產費用。欲回答供應商的這些問題與瞭解其生產控制的結構為何，我們發展出簡易的數量模型作評估與分析。模型採用動態規劃，多期的模式 (Dynamic Programming with finite time periods)，而在每一期的期初，供應商將依照收邊之前已生產的剩餘數量，來決定當期該生產多少產品。具有道德的買家會根據供應商的生產數量來決定購買的數量，此假設是根據實務上買家會鼓勵賣家多提供這樣的商品給他們，以善盡社會責任與同時協助供應商願意做這樣的轉型。模型一是設定為道德的供應商只可提供原物料與零組件給具有道德意識的買家，此假設是因為有時這類買家願意負較高的採購價格 (high premium)，其同時也希望商品只單獨提供給他們，即 exclusivie 的商品。我們分析其數學結構，發現生產的數量跟手邊庫存無關，反而跟所在的時間點有關。如時間是在整個 time periods 的初期，供應商會較保守生產少數量，但越靠近末期，供應商會增加生產數量。模型二提供了一個廣泛的設定，道德的供應商不只可提供原物料與零組件給具道德意識的買家，同時可將剩餘的數量提供給以價制量的買家。模型分析後發覺在傳統作業管理領域所謂的 up-to-level policy 存貨模型這樣的概念，並不適用於我們的議題。有趣的事，對這樣道德行為所產出的產品，市場機會的擴展吸引度將於供應商的生產行為息息相關。假如吸引度是相對低的，當供應商藉由之前留下的剩餘存貨越多，將減少本期的生產數量，但這樣的生產數量減少並不會與庫存的增加量成正比，這發現便是為何我們提及 order-up-to level 這樣的概念在本議題並不存在。同時當庫存水位高到一定地步時，供應商將於當期停止生產行為。 3. 會議中的討論 報告完後，與會者給予了兩個指教。 (1) 如何決定新的商業機會大小，是否有可能轉為 emperical study? 回答如以下: 因為目前實在沒有關於增加的市場需求實際數據與參數值為何，但未來文章將依照此一建議增加一個 numerical study section，來對各種重要的參數，如商業機會函數裡的參數，賣給道德意識的買家售價，與需求不確定的變異程度……等等參數以敏感度分析(sensitivity analysis)的方式來呈現，期能增加未來投稿的把握度。 (2) 供應商新的生產流程，是否有考慮潛在的風險危機? 回答如以下: 對於 entrepreneurship 的領域來說，對需求的不確定性與生產負責任的商品過程中的 yield rate，的確是值得關注的點。本文目前已經將需求的不確定性加入模型考量的一部分，而未來此研究也將專注生產過程的不良率因素於模型當中。

(27)

6

心得與建議:

此次會議發覺越來越多的 track 開始將永續議題作結合，包含在社會，經濟與環境議題上，也可看到。其中過往大家普遍以質化的方式來調查這個議題，但是現在很多學者也開始結合自身的領域與此一新議題，來發表新文章。而我這次的確也收集到了用量化方式(Game theory，EOQ 等等經濟與傳統作管領域模型)來針對不管是永續，綠色產品等等議題作探討的文獻，這將對我未來繼續在這塊領域上有很大的幫助。

攜回資料名稱及內容:

研討會論文 Proceeding.

(28)

科技部補助計畫衍生研發成果推廣資料表

日期:2016/01/29

科技部補助計畫

計畫名稱: 探索供應鏈架構下企業機會之分析：從採購與生產之角度計畫主持人: 陳立民計畫編號: 103-2410-H-004-107- 學門領域: 生產及作業管理

無研發成果推廣資料

(29)

103年度專題研究計畫研究成果彙整表

計畫主持人：陳立民計畫編號： 103-2410-H-004-107-計畫名稱：探索供應鏈架構下企業機會之分析：從採購與生產之角度成果項目量化單位備註（質化說明：如數個計畫共同成果、成果列為該期刊之封面故事...等）實際已達成數（被接受或已發表）預期總達成數（含實際已達成數）本計畫實際貢獻百分比國內論文著作期刊論文 2 3 100% 篇研究報告/技術報告 0 0 100% 研討會論文 1 1 100% 專書 0 0 100% 章/本專利申請中件數 0 0 100% 件已獲得件數 0 0 100% 技術移轉件數 0 0 100% 件權利金 0 0 100% 千元參與計畫人力（本國籍）碩士生 0 0 100% 人次博士生 0 0 100% 博士後研究員 0 0 100% 專任助理 0 0 100% 國外論文著作期刊論文 0 0 100% 篇研究報告/技術報告 0 0 100% 研討會論文 0 0 100% 專書 0 0 100% 章/本專利申請中件數 0 0 100% 件已獲得件數 0 0 100% 技術移轉件數 0 0 100% 件權利金 0 0 100% 千元參與計畫人力（外國籍）碩士生 0 0 100% 人次博士生 0 0 100% 博士後研究員 0 0 100% 專任助理 0 0 100% 其他成果（無法以量化表達之成果如辦理學術活動、獲得獎項、重要國際合作、研究成果國際影響力及其他協助產業技術發展之具體效益事項等，請以文字敘述填列。）　　無

(30)

成果項目量化名稱或內容性質簡述科教處計畫加填項目測驗工具(含質性與量性) 0 課程/模組 0 電腦及網路系統或工具 0 教材 0 舉辦之活動/競賽 0 研討會/工作坊 0 電子報、網站 0 計畫成果推廣之參與（閱聽）人數 0

(31)

探索供應鏈架構下企業機會之分析：從採購與生產之角度

科技部補助專題研究計畫成果報告

期末報告