Fig 1.
Conceptual graph for the service supply chain.
Fig 2.
The work flow graph for the ranking neural network.
Fig 3.
The framework of supplier portfolio selection algorithm.
Fig 4.
Features of supplier.
Fig 5.
Ranking Neural Network.
Fig 6.
Examples of suppliers in level I.
For the limitation of the space, we use some abbreviations. They are, P: Productivity, L: Location, F1: Quantity Flexibility, F2: Quality Flexibility, F3: Expectation Flexibility, QL: Quality Level, HRL: Human Resources Level, PL: Professional Level, IL: Informational Level, ES: Enterprize Scale, DA: Development Ability, FS: Financial State, CCr: Cooperative Credit, CI: Cooperative Intention, CCo: Cooperation Cost. Detailed data can be found in the supporting information (S1 Suppliers and Labeled Orders).
Fig 7.
Examples of the designed orders.
Each line shows the data example for some specific features. The features will be processed by our feature extraction tool programmed with Java programming language. Detailed data can be found in the supporting information (S1 Suppliers and Labeled Orders).
Fig 8.
Examples of the labeled suppliers portfolios.
Each line indicates a supplier portfolio which contains six suppliers in different categories. The last column is the degree which indicates the degree of each portfolio evaluated by the volunteers according to the possible profit for specific orders. Detailed data can be found in the supporting information (S1 Suppliers and Labeled Orders).
Table 1.
Results for the simulation experiments.